Full text of "Proceedings of the Annual Convention"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project
to make the world's books discoverable online.

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the
publisher to a library and finally to you.

Usage guidelines

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.

We also ask that you:

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liability can be quite severe.

About Google Book Search

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web

at |http : //books . google . com/

w
^P^

f^^.

THE GIFT OF MR. ALFRED GWYNNE VANDERBILT

UBRARY

RAILROAD BRANCH

YOUNG MEN'S CHRISTIAN ASSOCIATION

30» PARK AVENUE
NEW YORK

-'TV rt

- •> ^ /

^j^ m PROPERTY OP jm

ARTES SCIENTIA VERITAS

Digitized by

Google

Digitized by

Google

;

NATIONAL ASSQCMriON

• r •■ b

CEMENT UiitFt§:Tr.::

PROCEEDINGS

OF THE

Eighth Annual Convention

Held at Kansas City, Mo.,
March 11, 12, 13, 14, 15, 16, 1912

Volume VIII

EDITED UNDER DIRECTION OF THE PRESIDENT

BY THE SECRETARY

PUBLISHED BY THE ASSOCIATION
1912

Digitized by

Google

• • ••• • ••••••• •

•• :•: •• •!•• • • • • : •••:

• ••;• • ••• •• •

.••••••• ••••••

• • • • • • ••« • • • •««

! . • •• •'

• • • ••••••*

fraf^ station

s y^

Comaxan 1915 bt thb Ambbioak Concbbtb I mwitutb

The ABBOoiBtioii is not reaponaible, as a body, for the statements and opinions
advanced in iU publications.

Digitized by

Google

CONTENTS.

PAOfc

Personnel of Officers 6

Personnel of Sectional Committees 7

Personnel of Past Officers 9

Charter 10

By-Laws 11

Summary of Proceedings, Eighth Annual Convention 16

Annual Address by the President, European Practice in Concrete Con-
struction — Richard L. Humphrey 31

Report of the Committee on Reinforced Concrete and Building Laws —

A. E. Lindau, Chairman 61

Disciission 158

The Testing of Reinforced Concrete Buildings Under Load — W. A.

Slater 168

The Design of Concrete Flat Slabs— F. J. Trelease 218

Discussion 251

The Practical Design of Reinforced Concrete Flat Slabs — Sanford E.

Thompson 254

Discussion 274

The Design of Concrete Grain Elevators — E. Lee Heidenreich 277

Discussion 288

Report of the Committee on Measuring Concrete — ^Robert A. Cummings,

Chairman 290

Proposed Standard Methods for the Measurement of Concrete Work 301

Concrete Retaining Walls — John M. Meade 308

Discussion 311

Reinforced Concrete Piles — Robert A. Cummin^ 312

The Handling of Concrete in the Construction of the Panama Canal — S. B.

Williamson 326

Use of Concrete in the Fourth Avenue Subway, Brooklyn, N. Y. — Fred-
erick C. Noble 361

The Use of Reinforced Concrete in Hypochlorite Water Purification

Works— Walter M. Cross 372

Design and Construction of the Estacada Dam — H. V. Schreiber 376

Unit Cost of Reinforced Concrete for Industrial Buildings — C. S. Allen . . 400
Reinforced Concrete Convention Hall at Breslau, Germany — S. J. Trauer. 406
The Suitability of Concrete for Gas Holder Tanks— Herbert W. Ab-ich . . 412
Protection of Steel in Catskill Aqueduct Pipe Siphons — Alfred D. Flinn. . 424

A Fireproof School of Concrete — Theodore H. Skinner 444

Tlie Plresent Status of Unit Concrete Construction — ^James L. Darnell. . . 455
Discussion 469

(3)

24537 Digitized by Google

Contents.

PADS

Report of Committee on Specifications and Methods of Tests for Concrete

Materials — Sanford E. Thompson, Chairman 473

Aggregates for Concrete — William M. Kinney 486

Discussion 497

Field Inspection and Testing of Concrete — G. H. Bayles 501

Comparative Tests of the Strength of Concrete in the Laboratory and in

the Field— Rudolph J. Wig 522

Discussion 526

The Necessity for Field Tests of Concrete — Fritz X'on Emperger 530

Discussion 537

Report of the C'ommittee on Treatment of Ccmcrete Surfaces — L. C.

Wason, Chairman 539

Discussion 547

Cement Coatings — F. J. Morse 552

Discussion 561

Review of the Present Status of Iron Portland Cement — P. H. Hates. . . . 566

Marine or Iron Ore (Vments — Herman E. lirovvn 578

Iron Ore Cement — Arthur E. Williams 597

Discussion on Iron Ore Cement 613

Flat Slab Concrete Bridges— William H. Finley 616

Concrete Highway Bridges — Walter Scott Gearhart 621

Concrete Bridges — Daniel B. Luten 631

Discussion on Concrete Bridges 641

Report of Committee on Roadways, Sidewalks and Floors — C. W. Boyn-

ton, Chairman 644

Standard Specifications for Portland Cement Sidewalks 645

Standard Specifications for Concrete Roads and Street Pavements 651

Standard Specifications for Concrete Curb and C^oncret(> Curb and (Gutter 658

Standard Specifications for Plain Concrete Floors 665

Standard Specifications for Reinforced Concrete Floors 671

Discussion on Concrete Floors 676

An Improved Concrete Pavement — E. W. Groves 683

Cement Paving as Constructed at Mason City, Iowa — F. P. Wilson. . . 689

Discussion on Concrete Roads 694

Report of the Committee on Building Blocks and Cement Products —

P. H. Hudson, Chairman 699

Recommended Practice for Plain Concrete Drain Tile 700

Recommended Practice for Concrete Architectural Stone, Building

Block and Brick 703

Standard Specifications for Concrete Architectural Stone, Building Block

and Brick 707

Standard Building Regulations for the Use of Concrete Architectural

Stone, Building Block and Brick 710

Method of- Testing Cement Pipe — Arthur N. Talbot and Duflf A. Abrams 713
Advantages and Durability of Cement Sewer Pipe — Gustave Kaufman. . 720
The Manufacture and Use of Cement Drain Tile — C^harles E. Sims 727

Discussion 732

Digitized by

Google

Contents.

PAGE

Modern Methods of Manufacturing Concrete Products — Robert F.

Havlik 740

Notes on Reinforced Concrete Telegraph Poles — George Gibbs 757

Concrete Fence Posts — W. J. Towne 765

Concrete Fence Posts — L. J. Hotchkiss 766

Discussion 774

Report of Executive Board 779

Minutes of Meetings of the Executive Board 786

Subject Index 796

Author Index 804

List of Publications 809

Table of Contents 809

Price List 818

PLATES

The Handling of Concrotc in the Construction of the Panama Canal-
Williamson.

I. Fig. 1. — Plan and Cross-section Gatun Locks 328

II. Fig. 2.— Concrete Handling Plant, Gatun Locks 328

Fig. 3.— Wall Forms, Gatun Locks 328

III. Fig. 6. — (>)ncrcte Handling Plant, Mirafloroa Ix)cks 340

IV. Fig. 7. — Details, C>)ncretc Handling Plant, Miraflores Locks. . . 342
V. Fig. 10. — Forms for Wall, Pedro Miguel and Miraflores I>ocks. . . . 344

VI. Fig. 11.— Handling Plant, Pedro Miguel I^cks 346

VII. Fig. 13. — Material Handling Cranes, Pedro Miguel Ivocks 346

Digitized by

Google

LIST OF OFFICERS

THE NATIONAL ASSOCIATION OF CEMENT USERS

1912.

PRESIDENT.

RICHARD L. HUMPHREY.

FIRST VICE-PRESIDENT.

EDWARD D. BOYER.

SECOND VICE-PRESIDENT.

ARTHUR N. TALBOT.

THIRD VICE-PRESIDENT.

EDWARD S. LARNED.

FOURTH VICE-PRESIDENT.

IRA H. WOOLSON.

SECRETARY.

EDWARD E. KRAUSS.

TREASURER.

HENRY C. TURNER.

SECTIONAL VICE-PRESIDENTS.
(See page 7.)

Digitized by

Google

SECTIONAL COMMITTEES.

(The Chainnen are Vice-Presidents of the Association.)

BUILDING BLOCKS AND CEMENT PRODUCTS.
P. S. Hudson, Chairman,

C. K. ABPy ROBERT V. HAYUK,

P. H. ATWOOD, CHARLES D. WATSON.

EXHIBITION.
H. S. DoTLB, Chairman,

B. F. AFFI4ECK, W. J. ROSBBERRT, JR.,

W. H. BURKE, P. AUSTIN TOMES.

FIKEPROOFING.
Rudolph P. Miller, Chairman,

EDWIN CLARK, JOHN STEPHEN SEWELL,

W. C. ROBINSON, IRA H. WOOLSON.

INSURANCE.
William H. Ham, Chairman,

A. L. JOHNSON, F. W. MOSES,

W. H. MERRILL, J. P. H. PERRY.

MEASURING CONCRETE.
Robert A. Cummings, Chairman.

L. H. ALLEN, CHARLES DERLETH, JR.,

ROBERT ANDERSON, THOMAS M. VINTON.

NOMENCLATURE.
Peter Gillespie, Chairman,

E. p. GOODRICH, F. B. TURNEAURE,

E. J. MEHREN, FRANK C. WIGHT.

REINFORCED CONCRETE AND BUILDING LAWS.
Alfred E. Lindau, Chairman,

W. p. ANDERSON, ARTHX7R N. TALBOT,

E. J. MOORE, 8ANF0RD B. THOMPSON.

ROADWAYS, SIDEWALKS AND FLOORS.
C. W. BoYNTON, Chairman.

A. G. BIRNIE, C. R. MILLER,

i, B. LANDFPXD, A. B. SN0DGRAB8,

(7)

Digitized by

Google

Sectional Committees.

SPECIFICATIONS AND METHODS OF TESTS P'OR CONCRETE

MATERIALS.

Sanpord E. Thompson", Chairman.

CLOYD M. CHAPMAN, RUSSELL 8. GRBENMAN,

WILLIAM B. FULLER, ARTHUR N. TALBOT.

TREATMENT OF CONCRETE SURFACES.
Leonard C. Wason, Chairman.

CLOYD M. CHAPMAN, EMILE O. PERROT,

ALFRED HOPKINS, H. H. QUIMBY.

EDUCATION.

Logan Waller Page, Chairman.
Percy H. Wilson, Secretary.
A. C. True. Morris Mbtcalf.

William G. Hartranft. A. Moyer.

W. A. Holman. J. P. H. Perry.

Agricultural Experiment Stations.

Alabama — J. F. Duogar.
Arizona — Representative not yet

appointed.
Arkansas — Martin Nelson.
California — LeRoy Anderson.
Colorodo — Alvin Keyser.
Connecticut — Charles A. Wheeler.
Delaware — Harry Hayward.
Florida — ^J. J. Vernon.
Georgia — LeRoy C. Hart.
Idaho — W. L. Carlyle.
Illinois — E. A. White.
Indiana — A. T. Wiancke.
Iowa — J. B. Davidson.
fiansas — W. W. Jardjnk.
Kentucky — M. A. Scovell.
I^uisiana — W. R. Dodson.
Maine — George E. Summons.
Maryland — W. T. L. Taliaferro.
Massachusetts — Wiluam D. Hurd.
Michigan — R. S. Shaw.
Minnesota — John T. Stewart.
Mississippi — W. L. Hutchinson.
Missouri — F. B. Mumford.
Montana — H. B. Bonebright.

Nebraska — L. W. Chase.
Nevada — Representative not yet

appointed.
New Hampshire — F. W. Taylor.
New Jersey — ^J. G. Lipman.
Neiv Mexico — F. L. Bixby.
New York—n. W. Riley.
North Carolina — C. L. Newman.
North Dakota — ^J. H. Shepperd.
Ohio — H. C. Hamsower.
Oklahoma — O. O. Churchill.
Oregon — H. D. Scudder.
Pennsylmnia — Frank D. Gardener.
Rhode Island — G. E, Adams,
South Carolina — W. R, Perkjns.
South Dakota — A, N. Hume.
Tenntissee — C, A. Moores,
Texas — S. S. McMjllant.
IJtah—F, S. Harris.
Vermont — J. W. Elliot.
Virginia — H. L. Price.
Washington — O. L. Waller.
West Virginia — I. S. Cook.
Wisconsin — Charles A. Ocock.
Wyoming — J. C. F|tterer.

Digitized by

Google

PAST OFFICERS.

President.

1905 John P. Given.

(Presiding Officer First Convention.)

1905-11 Richard L. Humphrey.

First Vice-President.

1905 a. l. goetzmann.
1906-9 Merrill Watson.
1909-11 Edward D. Boyer.

Second Vice-President.

1905-6 John H. Fellows.
1907-10 M. S. Daniels.
1911 Arthur N. Talbot.

Third Vice-President.

1905 H. C. QuiNN.
1906-7 0. U. Miracle.
1908 S. B. Newberry.
1909-11 E. S. Larned.

Fourth Vice-President.

1905-7 A. Monsted.
1908-9 George C. Walters.
1909-10 F. A. NoRRis.
1911 Ira H. Woolsen.

Treasurer.

1905 A. S. J. Gammon.
1906^11 H. C. Turner.

Secretary.

1905-6 Charles C. Brown.

1907 W. W. Curtis.

1908-9 George C. Wright.

1910 Edward E. Krauss (Acting)

1911 Edward E. Krauss,

(9)

Digitized by

Google

CHARTER

THE NATIONAL ASSOCIATION OF CEMENT USERS.

KNOW ALL MBN BY THBSB PRBSBNTS, That we, the under-
signed, all of whom are citizens of the United States, and a majority of
whom are residents of the District of Columbia, have associated our-
selves together for the purpose hereinafter set forth and desiring that
we may be incorporated as an Association under sub-chapter three (3) of
the Incorporation Laws of the District of Columbia, as provided in the
Code of Law of the District of Columbia, enacted by Congress and ap-
proved by the President of the United States, do hereby certify:

/• Nmm€. The name of the proposed corporation is "The National
Association of Cement Users."

2. Term oi Bxiateace. The existence of the said corporation shall
be perpetual.

3. OtiectMm The particular business and objects of the said corpora-
tion shall be to disseminate information and experience upon and to
promote the best methods to be employed in the various uses of cement by
means of convention, the reading and discussion of papers upon materials
of a cement nature and their uses, by social and friendly intercourse at
such conventions, the exhibition and study of materials, machinery and
methods and to circulate among its members by means of publications the
information thus obtained.

4. iacorpormton. The number of its managers for the first year
shall be fifteen.

la WItaeMB Whereof, we have hereunto set our hands and seals this
fourteenth day of December, A. D. 1906.

RICHARD L. HUMPHREY, (Seal)

JOHN STEPHEN SEWELL, (Seal)

S. S. VOORHEES. (Seal)

Office of Recorder of Deeds,
District of Columbia.
This is to certify that the foregoing is a true and verified copy of a
Certificate of Incorporation, and of the whole of such Certificate as re-
ceived for record in this office at 9 : 49 A. m., the 19th day of December,
A. D. 1906.

In testimony whereof I have hereunto set my hand and affixed the
teal of this office, this 20th day of December, A. D. 1906.
(Signed) R. W. DUTTON,

Deputy Recorder of Deeds,

LHstrict of Columbia

(10)

Digitized by

Google

BY-LAWS.
Article I.

MEMBERS.

Section 1. Any person engaged in the construction or
maintenance of work in which cement is used, or qualified by
business relations or practical experience to co-operate in the pur-
poses of this Association, or engaged in the manufacture or sale
of machinery or supplies for cement users, or a man who has
attained eminence in the field of engineering, architecture or
applied science, is eligible for membership.

Sec. 2. A firm or company shall be treated as a single mem-
ber.

Sec. 3. Any member contributing annually twenty or more
dollars in addition to the regular dues shall be designated and
listed as a Contributing Member.

Sec. 4. Application for menibership shall be made to the
Secretary on a form prescribed by the Board of Direction. The
Secretary shall submit monthly or oftener if necessary to each
member of the Board of Direction for letter ballot a list of all
applicants for membership on hand at that time with a statement
of the qualifications, and a two-thirds majority of the members of
the Board shall be necessary to an election.

Applicants for membership shall be qualified upon notification
of election by the Secretary by the payment of the annual dues,
and unless these dues are paid within 60 days thereafter the elec-
tion shall become void. An extract of the By-Laws relating to
dues shall accompany the notice of election.

Sec. 5. Resignations from membership must be presented
in writing to the Secretary on or before the close of the fiscal year
^pd shall be acceptable provided the dues are paid for that year,

Digitized by

Google

12 By-Laws.

Article II.

OFFICERS.

Section 1. The officers shall be the President, two Vice-
Presidents, six Directors (one from each geographical district),
the Secretary and the Treasurer, who, with the five latest living
Past-Presidents, who continue to be members, shall constitute
the Board of Direction.

Sec. 2. The Board of Direction shall, from time to time,
divide the territory occupied by the membership into six geographi-
cal districts, to be designated by numbers.

Sec. 3. The terms of office of the President, Secretary and
Treasurer shall be one year; of the Vice-Presidents and the Direct-
ors, two years. Provided, however, that at the first election after
the adoption of this By-Law, a President, one Vice-President,
three Directors and a Treasurer shall be elected to serve for one
year only, and one Vice-President and three Directors for two
years; provided, also, that after the first election a President, one
Vice-President, three Directors and a Treasurer shall be elected
annually.

The term of each officer shall begin at the close of the Annual
Convention at which such officer is elected, and shall continue for
the period above named or until a successor is duly elected.

A vacancy in the office of President shall be filled by the senior
Vice-President. A vacancy in the office of Vice-President shall
be filled by the senior Director.

Seniority between persoas holding similar offices shall be
determined by priority of election to the office, and when these
dates are the same, by priority of admission to membership;
and when the latter dates are identical, the selection shall be made
by lot. In case of the disability or neglect in the performance of
his duty, of any officer of this Association, the Board of Direction
shall have power to declare the office vacant. Vacancies in any
office for the unexpired term shall be filled by the Board of Direc-
tion, except as provided above.

Sec. 4. The Board of Direction shall appoint the Secretary;
it shall create such special committees as may be deemed desirable
for the purpose of preparing recommended practice and standards
concerning the proper use of cement for consideration by the

Digitized by

Google

By-Laws. 13

Association, and shall appoint a chairman for each committee.
Four or more additional members on each special committee
shall be appointed by the President, in consultation with the
Chainnan.

Sec. 5. It shall be the duty of the Board of Direction to
audit the accounts of the Secretary and the Treasurer before each
annual convention.

Sec. 6. The Board of Direction shall appoint a Committee
on Nomination of Officers and a Committee on Resolutions, to be
announced by the President at the first regular session of the
annual convention.

Sec. 7. There shall be an Executive Committee of the
Board of Direction, consisting of the President, the Secretary, the
Treasurer and two of its members, appointed by the Board of
Direction.

Sec. 8. The Executive Committee shall manage the affairs
of the Association during the interim between the meetings of
the Board of Direction.

Sec. 9. The President shall have general supervision of the
affairs of the Association. He shall preside at the Annual Con-
vention, at the meetings of the Board of Direction and the Execu-
tive Committee, and shall be ex-officio member of all conmiittees.

The Vice-Presidents in order of seniority shall discharge the
duties of the President in his absence.

Sec. 10. The Secretary shall perform such duties and fur-
nish such bond as may be determined by the Board of Direction.

Sec. 11. The Treasurer shall be the custodian of the funds
of the Association, shall disburse the same in the manner prescribed
and shall furnish bond in such sum as the Board of Direction may
determine.

Sec. 12. The Secretary shall receive such salary as may be
fixed by the Board of Direction.

Article III.

MEETINGS.

Section 1. The Association shall meet annually. The time
and place shall be fixed by the Board of Direction and notice of
this action shall be mailed to all members at least thirty days
previous to the date of the Convention.

Digitized by

Google

14 bT-tiAW&.

Sec. 2. The Board of Direction shall meet immediately
after the Convention at which it was elected, effect organization
and transact such business as may be necessary.

Sec. 3. The Board of Direction shall meet at least twice
each year. The time and place to be fixed by the Executive Com-
mittee.

Sec. 4. A majority of the members shall constitute a
quorum for meetings of the Board of Direction and of the Executive
Committee.

Article IV.

DUES.

Section 1 The fiscal year shall commence on the first
of July and all dues shall be payable in advance.

Sec. 2. The annual dues of each member shall be five dollars
($5.00).

Sec. 3. Any person elected after six months of any fiscal
year shall have expired, need pay only one-half of the amount
of dues for that fiscal year; but he shall not be entitled to a copy
of the Proceedings of that year.

Sec. 4. A member whose dues remain unpaid for a period of
three months shall forfeit the privilege of membership and shall
be officially notified to this effect by the Secretary, and if these
dues are not paid within thirty days thereafter his name shall be
stricken from the list of members. Members may be reinstated
upon the payment of all indebtedness against them upon the books
of the Association.

Article V.

RECOBfBfENDED PRACTICE AND SPECIFICATIONS.

Section 1. Proposed Recommended Practice and Specifica-
tions to be submitted to the Association must be mailed to the
members at least thirty days prior to the Annual Convention, and
as there amended and approved, passed to letter ballot, which shall
be canvassed within sixty days thereafter, such Recommended
Practice and Specifications shall be considered adopted unless at
least ten per cent, of the total membership shall vote in the
negative.

Digitized by

Google

Akticle VI.

AMENDMENT

Section 1. Amendments to these By-Laws, signed by at
least fifteen members, must be presented in writing to the Board of
Direction ninety days before the Annual Convention and shall be
printed in the notice of the Annual Convention. These amend-
ments may be discussed and amended at the Annual Convention
and passed to letter ballot by a two-thirds vote of those present.
Two-thirds of the votes cast by letter ballot shall be necessary
for their adoption.

Digitized by

Google

SUMMARY OF THE PROCEEDINGS OF THE EIGHTH
ANNUAL CONVENTION.

First Session, Monday, March U, 1912, 8 p. m.

The Convention was called to order by the President, Richard
L. Humphrey.

John Lyle Harrington, Past President, Engineers' Club of
Kansas City, delivered an address of welcome on behalf of the
Engineering Interests as follows:

In the dark ages of industry which extended well into the last century,
it was the custom for every member of a craft, trade or profession to guard
jealously and to keep closely secret every item of knowledge he or his associates
had, in order to secure to himself the whole advantage of it. In addition to
patenting an invention, it was common to keep secret every possible detail
of the processes employed, and many businesses were based wholly upon
secret formulae which were closely held by members of a firm or family, often
handed down from father to son, and the utmost precautions were taken to
ensure that employees and even associates, as well as competitors, actual and
potential, should be kept in ignorance of the methods employed or discoveries
made. Even up to the present day it is dangerous in some e.stablishments
for an employee to ask too many questions regarding the methods of manu-
facture or materials he employs in his work, and in certain lines of manufacture
a considerable remnant of this old secrecy remains. Here and there the
possessor of a formula has, like the old alchemist, at the right moment dropped
his fluid into a molten metal or added his mite to the production of important
materials and kept the secret and profits thereof to himself.

But early in the last century the civil engineers of Great Britain met and
formed an institute for the purpose of disseminating the knowledge acquired
by its individual members; and in the latter half of the century the engineers
of this country came to appreciate the advantages of co-operation. The
first such organizations were few in number and comprehensive in scope,
but gradually important groups working in special lines came to feel that the
interests of the broad general organization were too varied to permit adequate
consideration of the matters which specially occupied their attention and so
they split off and organized societies of more limited .scope. With the enor-
mous development of industries based on the appliinl sciences special interests
so increased in value and importance that the benefits to be derived from close
association and active discussions of men engaged in them came to be generally
understood, and group organizations grew apace.

(10)

Digitized by

Google

Summary of Proceedings, Eighth Convention. 17

In the course of time, manufacturers began to understand that by keep-
ing to themselves knowledge of their s|x;cialities they encouraged like action
on the part of their competitors — and each member of that branch of industrj'
developed internally only. This limited the development of all and circum-
scribed the field of the industry. Gradually it became evident that the
advancement of the industry resulting from the dissemination of knowledge
of it among those engaged in it more than compensated for the advantage
secured even by the most successful operation under the secret methods. As
soon as this fact came to be fully realized the organization of special societies
for special purposes multiplied rapidly. They reacted upon the industries
and the accomplishments of one man so pointed out possibilities and so
stimulated others that development has increased in geometric ratio.

Among these later organizations developed to special interests, the
National Association of Cement Users has come to occupy an important
place.

Cement in some form has been in use for many centuries, so many that
its origin is lost in antiquity; but the development of Portland cement and
the industries and types of construction dependent upon its use are of com-
paratively recent date. It is hard to believe that less than twenty years ago
Congress gravely questioned the existence in this country of materials essential
to the manufacture of first-class Portland cement, and debated whether, in
view of that condition, it would be justifiable to impose so much duty on
imported cement as would induce its manufacture in this country. Yet hist
year we produced nearly 78,(XX),(XX) barrels, which, at the existing low price
prevailing, brought nearly $(>8,(XX),(XX) at the mills, and the value of con-
structions in which it wjis used probably reached nearly $500,000,000.

It is well, therefore, that the users of cement, the men who are respon-
sible for construction of cement and concrete work worth one-half billion
dollars per annum, should organize themselves into an association and meet
to advance their special interests by the discussion of the work they are doing
and the experiences they have gained. And while it is natural and right that
interest and enthusiasm for their work should lead chiefly to the exposition
of the successes achieved, it is quite as important that the defects found in the
materials, in the constructions and in the methods employed should be
exposed and discussed, for we learn as much — often more, if we are wise —
from our failures as from our successes. And it is important that the limita-
tions of materials and their uses be fully understood and bad results thus
guarded against, and the character of the practices on th<i whole thus improved.
The effort to meet the ever-pressing demand for cheap construction leads to
many failures and to much consequent damage to the industry. To increase
the stresses or reduce the quality of concrete in order to enable it to compare
favorably with wood in first cost is an unwarranted, but far too common
practice. If in any instance its many superior qualities do not justify the
greater expenditure for good concrete safely stressed, then the cheaper mate-
rials should be employed. Wilful, foolhardy rLsks are responsible for some of
the failures, but ignorance of all conditions governing concrete construction

Digitized by

Google

18 Summary op t*ROCEia)iNGS, Eighth ConvenHok.

is by far the larger factor, and it is the duty and purpose of this organization
to expose bad construction as it is to urge good methods and good workmanship.
Concrete construction is peculiarly liable to failure through both ignorance
and carelessness. Through ignorance because its use in the simpler tjrpes of
construction leads men to believe its employment universally simple; through
carelessness because it is so amenable to the employment of unskilled labor
which needs, but does not always receive, thorough and careful supervision.
I well remember the county engineer who was confident he had made all the
plans necessary for the construction of a reinforced concrete bridge when he
had made a picture on which he had given general dimensions and shown the
location of some reinforcement.

Failure to recognize the need of careful inspection of materials and super-
vision of workmanship is perhaps responsible for the larger portion of failures
of concrete structures. Steel is manufactured by skilled men working under
able superintendence and so continuously employed that they come to under-
stand the processes thoroughly; whereas concrete is conmionly manufactured
by an itinerant, common laborer and often directed by a foreman whose
only qualification is his ability to handle men. Constructions of wood,
brick or stone may be fairly well inspected at any stage or after completion,
but the character of concrete construction may not be determined until
after the forms are removed, and remedying defects then disclosed is always
difficult, often impossible. Too early removal of the forms, thus placing upon
the concrete stresses which it is yet unfitted to bear; carelessness and imwise
placing of reinforcement; inadequate provision for expansion and contrac-
tion due to changes of temperature; weak and leaky forms; inadequate
tamping; loose methods of depositing concrete, both in air and under water;
excessive dependence in designing upon empirical methods and fallacious
load tests; these and many other difficulties must be guarded against and
overcome. Much disappointment is certainly in store for the owners of
staff and concrete buildings who have not fully appreciated the difficulties
in securing construction which will resist the weather. There are many other
difficulties which must be guarded against and overcome.

As the use of concrete becomes more general, more attention is given to
the sightliness of structures built of it, but too often the efforts in this line
are misdirected. It should be clearly recognized that this material is capable
of excellent treatment peculiar to itself, and that it is an error to try to
make it resemble other materials. The old Spanish structures of southwestern
states and Mexico are beautiful because they are true. They were designed
frankly to be built of concrete and have forms and finish suitable for that
material. It may be that with the wider use and fuller development of con-
crete construction we may improve upon the work of these Spaniards, but
we have not by any means equalled it as yet. The endeavor to meet the needs
of the material has too frequently led to the adoption of extreme and grotesque
forms and finishes which soon weary or offend.

We are gradually improving the finish of our concrete structures and
recent work in this line is especially promising, but much remains to be done
both for the appearance and for the weathering quality of concrete and staff

Digitized by

Google

SUMMART OP PROCEEBtNGft, ElGHTH CoNVfcNTtOI^. 19

^surfaces. Good finish adds to the cost, but the extra expense is surely jus-
tified.

The gfeat advantages of concrete, durability, loW cost as compared with
other equally durable materials, ease of handling, ease of molding to the forms
desired, resistance to fire, strength and homogeneity) adaptability to many
uses, both with and without reinforcement, resistance to acids and decay^
attractiveness of appearance and the wide distribution of its constituent
materials, improvements in methods of construction, and the steadily
increased scarcity and cost of good timber ensure continual increase in the
use of concrete. But perhaps the greatest argument in favor of fireproof
concrete buildings is the enormous fire loss this country sustains through
the 'excessive use of wood and the equally great cost of fire insurance. Our
losses by fire have become a national disgrace.

The influence of this Association should be very great in combating the
tendency of some constructors to secure business and profit by reducing the
quality of work. This course surely militates against the interest of cement
users as a whole and it brings their work as a whole into disrepute. This was
clearly exemplified recently at Los Angeles where the city engineer recom-
mended creosoted instead of reinforced concrete piles for dock construction
and was able to justify his position by citing about a dozen failures of docks
which were supported on reinforced concrete piles. And generally speaking,
any improper use of the material of construction affects adversely the inter-
ests of all who engage in the use of that material. There is, therefore, a large
opportunity for this Association to benefit greatly the interests of its members
and to benefit the country at large by compelling a high standard of ability
and integrity among cement users.

It is thus apparent that this Association has met to deal with subjects
of the largest consequence and the effects of its discussions will be to improve
throughout this country and other countries the practices of concrete design
and construction. The work of this Association is, in no inconsiderable
measure, the work of the Engineer. He shares largely in the work and the
responsibilities and the benefits of this convention; hence it is with pleasure
and cordial good will that on behalf of the Engineers of Kansas City, I extend
a hearty welcome to the National Association of Cement Users.

An address of welcome on behalf of the Concrete Interests
was made by F. W. Pratt, President of the Union Bridge and
Terminal Railroad Company.

The President responded:

The Convention has this year been fortunate in haying two eminent
engineers express thoughts that will be of good service to the Association
in its future work. I am sure that you all join me in reciprocating the
good wishes which Mr. Harrington and Mr. Pratt have extended to us, and
in the hope that our deliberations may prove of interest and value.

Digitized by

Google

20 Summary of t^ROckEDiNGs, Eighth Convention.

The following committees of the Convention, appointed by
the Executive Board, were announced by the President:

Committee on Nomination of Officers:

E. J. Moore, Chairman^ New York, N. Y.
D. A. Abrams, Ames, Iowa.

John L. Conzelman, St. Ix)uis, Mo.

B. F. Lippold, Chicago, III.

Ij. T. Sunderland, KanHa.s CMty, Mo.

Committee on Resolutions:

Rudolph P. Miller, Chairman, New York, N. Y.
P. n. Bates, PittsburKh, Pa.
Allen Brett, New York, N. Y.

F. L. Williamson, Kansjus C'ity, Mo.
Perry H. Wilson, Philadelphia, Pa.

A paper on *'Th(» Use of Reinforced Concrete in Hypochlorite
Water Purification Works" was read by Walter M. Ooss.

(leorge E. Tebbetts presented a paper on *'The Use of Con-
crete in the New Union Station at Kansas City, Mo.," which
wjis followed l)y a discussion.

The meeting adjourncMl until Tuesday at 10.30 A. m.

Tuesday, March 12, 1912, 10.00 a. m.

Meeting of the Sections on Mcnisuring ('oncrete. Nomen-
clature, and Specifications and Methods of Tests for Concrete
Materials.

President Richard L. Humphrey in the chair.

The meeting took the form of a discussion on the deposition
of mortar with compressed air, its effectiveness and various
methods of application.

Second Session, Tuesday, March 12, 1912, 10.30 a. m.

President Richard L. Humphrey in the chair.

The report of the Committee on Specifications and Methods
of Tests for ('oncrete Materials was, in the absence of the Chair-
man, Sanford E. Thompson, presented by Cloyd M. Chapman.

Wm. M. Kinney read a paper on "Aggregates for Concrete,"

Digitized by

Google

Summary of Proceedings, Eighth Convention. 21

which was followed by a Topical Discussion on Concrete Aggre-
gates.

In the absence in Europe of the Chairman, Robert A.
Cummings, Edw. D. Boyer presented the report of the Com-
mittee on Measuring Concrete, which was followed by a dis-
cussion. The Proposed Standard Method of Measuring Concrete
was referred back to the Committee, with instructions for a
revision in the light of the discussion and report at the next
annual Convention.

Third Session— Tuesday, March 12, 1912, 8.00 p. m.

President Richard L. Humphrey in the chair.
The annual address of the President, entitled **The Use of
Concrete in Eurojx*,*' was delivered by Richard L. Humphrey.
The following papers were then read and discussed:

"The Design and Construction of the Hollow Reinforced
Concrete Dam of the Portland Railway Light and
Power Company," by Herman V. Schreiber.

"Cement Coatings in Color,'' by F. J. Morse.

The report of the Committee on Concrete Surfaces was
presented by the C'hairman, L. C. Wason. On motion the pro-
posed Standard Method for Tests of Waterproofing was referred
to the Committee on Specifications and Methods of Tests for
Concrete Materials, for report. Consideration of the proposed
Standard Specification for Portland Cement Stucco was deferred
imtil a later session. The changes in the last general report of
the Committee were approved.

The report of the Committee on Insurance was, in the
absence of the Chairman, Wm. H. Ham, read by title.

The meeting then adjourned until Wednesday at 9.30 a. m.

Fourth Session — Wednesday, March 13, 1912, 9.30 a. m.

President Richard L. Humphrey in the chair.

In the absence of the author, Sanford E. Thompson, the
President read the paper entitled "The Practical Design of
Reinforced Concrete Flat Slabs."

Digitized by

Google

22 SUMliART OP PROCEEDINGS; ElGHTH CONVENTION.

The report of the Committee on Reinforced Concrete and
Building Laws was presented by the Chairman, Alfred E. Lindau,
and Arthur N. Talbot, and was followed by a discussion.

The President then presented the annual report of the
Executive Board, submitting a proposed revision of the By-Laws,
which were amended and ordered to letter ballot. The report
was approved.

The change of name of the Association was discussed and
on motion the matter was referred to the Executive Board, with
authority to effect a change in name if deemed expedient for the
best interests of the Association.

The Committee on Nomination of Officers presented the
following nominations, which were unanimously approved and
the Secretary instructed to cast the ballot for their election:

Presidenty Richard L. Humphrey, Philadelphia, Pa.
First Vice-President, Edward D. Boyer, Catasauqua, Pa.
Second Vice-President, Arthur N. Talbot, Urbana, 111.
Third Vice-President, Edward S. Lamed, Boston, Ma«s.
Fourth Vice-President, Ira H. Woolson, New York, N. Y.
Treasurer, Henry C. Turner, New York, N. Y.

The time and place of the Ninth Annual Convention was
referred to the Executive Board with power to act.
The meeting then adjourned imtil 3 p. m.

Fifth Session — ^Wednesday, March 13, 1912, 3.00 p. m.

President Richard L. Humphrey in the chair.
The following papers were read and discussed:

"The Testing of Reinforced Concrete Buildings under

Load," by W. A. Slater.
"The Design of Concrete Flat Slabs," by Frank J. Tre-
lease; in the absence of the author read by Alfred E.
Lindau.
"The Present Status of Unit Construction," by James L.
Darnell.
In the absence of the author, Theodore H. Skinner, the
paper on "A Fireproof School of Concrete," was read by title.
The meeting then adjourned until 8.00 p. m.

Digitized by

Google

SUMMABT OF PROCEEDINGS, ElGHTH CONVENTION. 23

Sixth Session — ^Wednesday, March 13, 1912, 8.00 p. m.

President Richard L. Humphrey in the chair.

The President introduced the Honorable Darius A. Brown,
Mayor, who extended an address of welcome to Kansas City as
follows:

Mayor Brown. — I am glad of the opportunity to appear before you and
express my appreciation of the fact that men should gather together in this
city for the purpose of discussing the matters which have brought you here.
We have heard about the Stone Age, the Wooden Age and the Iron Age, and
I think it is the consensus of opinion that we are now in the Concrete Age
and that all of the improved structures are coming to be made of this material.
We are also in the age of progress and advancement; we are in an age when
people have adopted the idea that if a thing is worth doing at all it is worth
doing well and that is the reason why in every branch of human industry,
in business, in commerce, in the sciences and in the professions, they gather
together periodically for the purpose of discussing the ways and means of better
doing the business in which they are engaged.

I am satisfied that the result of your deliberations will not only be of
benefit to you in your particular business but will be of benefit to the com-
munity in which you live and benefit to the American people as a whole.
I hope one of the results of your deliberations will be that the use of concrete
will become so perfected that it will not only give us more comfortable build-
ings and better and more ornamental structures but will decrease the cost of
those materials to the people; because that is one of the great problems to be
solved.

On behalf of the people of Kansas City I want to extend to you a very
cordial welcome and trust that you will not only be benefited by your delibera-
tons but that you will have some pleasure while staying in our city. I thank
you.

The President responded:

I am sure we all appreciate the welcome that has been extended by His
Honor. A city of this size in whose immediate vicinity there is a production
of Portland cement of about one-tenth of that of the entire country, is a
good place in which to hold deliberations of this character. In accepting
the hospitality of this city we do so with a feeling that we will be benefited.
Our conventions have been held heretofore east of the Mississippi River,
and I believe that the mingling of the eastern and western ends of this
great country cannot help but be beneficial to all. I know you will join me
in extending to Mayor Brown hearty thanks for his welcome.

A paper on "The Use of Reinforced Concrete on the Wabash
Railroad," was presented by A, 0. Cunningham,

Digitized by

Google

24 Summary of Proceedings, Eighth Convention.

The report of the Committee on Treatment of Concrete
Surfaces was taken up and on motion the proposed Standard
Specification for Portland Cement Stucco was received as in-
formation and the Committee instructed to confer with other
Associations, in order to reach an agreement as to a specifica-
tion to be presented at the next Convention.

The following papers were then read and discussed:

"The Design and Construction of a Reinforced Concrete
Dome, 220 Foot Span," by S. J. Trauer; in the
absence of the author read by the President.
'*Tho Design of Concrete (irain Elevators," by E. Ja'v

Heidenreich.
'*The Suitability of Concrete for (liis Holder Tanks/' by
Herbert W. Alrich; in the absence of the author
presented by the President.
"The Necessity of Field Tests for Concrete," by Fritz E.
Von Emperger; in the absence of the author pre-
sented by the Secretary.
The meeting adjourned until Thursday at 10.30 a. m.

Thursday, March 14, 1912, 10.00 a. m.

Meeting of the Section on Treatment of Concrete Surfaces.
President Richard L. Humphrey in the chair.
The meeting took the form of a topical discussion on the
coloring of concrete surfaces, contraction, dusting of floors, etc.

Seventh Session— Thursday, March 14, 1912, 10.30 a. m.

President Richard L. Humphrey in the chair.
The following papers were read and discussed:

"Concrete Highway Bridges," by William Scott Gearhart.

"Concrete Bridges," by Daniel B. Luten.

"Flat Slab Bridges," by William H. Finley; in the absence
of the author read by the President,
The meeting adjourned until 8.00 p. m,

Digitized by

Google

Summary of Proceedings, Eighth Convention. 25

Eighth Session — ^Thursday, March 14, 1912, 8.00 p. m.

President Richard L. Humphrey in the chair.
The paper on "The Necessity for Good Roads/' by Logan
Waller Page, was in the absence of the author presented by E. L.
Eldredge.

The following papers were then presented :

"The Necessity of National Aid in Good Roads," by H. C.

Gilbert.
"Cement Paving as Constructed at Mason City, Iowa," by

F. P. Wilson.
"An Improved Concrete Pavement," by E. W. Groves.
The report of the Committee on Roadways, Sidewalks and
Floors, was presented by the Chairman, C. W. Boynton, and
the following action taken:

Proposed revisions of the Standard Specifications for Con-
crete Road and Street Pavements ordered to letter ballot.

The proposed revision of the Specifications on Sidewalks,
Curb and Gutter, and the proposed new Specifications for Plaiin
and Reinforced Concrete Floors, were considered and referred
to the Committee for report at a later session.

The meeting then adjourned until Friday at 10.30 a. m.

Friday, March 15, 1912, 9.00 a. m.

Meeting of the Section on Roadways, Sidewalks and Floors;
C. W. Boynton, Chairman of the Section, in the chair.

The meeting took the form of a discussion on concrete roads,
concrete floors, the prevention of dusting of floors, concreting in
freezing weather, etc.

Ninth Session— Friday, March 15, 1912, 10.30 a. m.

President Richard L. Humphrey in the chair.
The following papers were read and discussed:

"Concrete Fence Posts," by L. J. Hotchkiss.

"The Design of Reinforced Concrete Retaining Walls," by
John M. Meade.

Digitized by

Google

26 Summary of Proceedings, Eighth Convention.

"Advantages and Durability of Cement Sewer Pipe/' by
Gustave Kaufman; in the absence of the author
presented by the President.
"Methods of Testing Cement Pipe," by Duff A. Abrams;
in the absence of the author presented by W. A. Slater.
"The Manufacture and Use of Cement Drain Tile," by
Charles E. Sims.
The report of the Committee on Cement Products and Build-
ing Blocks was in the absence of the Chairman, Percy S. Hudson,
presented by Clarence K. Arp. On motion the Proposed Standard
Recommended Practice for Cement Tile was referred to letter
ballot.

The Committee on Roadways, Sidewalks and Floors, C. W.
Boynton, Chainnan, reported back the matters referred to it
and on motion the following were referred to letter ballot:

Proposed Revisions of the Standard Specification for

Portland Cement Sidewalks.
Proposed Revisions of the Standard Specifications for

Portland Cement Curb and Curb and Gutter.
Proposed Standard Specifications for Plain Concrete

Floors.
Proposed Standard Specifications for Reinforced Concrete
Floors.
The report of the Committee on Nomenclature was in the
absence of the Chairman, Peter Gillepsie, presented by Frank C.
Wight and was on motion accepted as information.
The Committee on Education reported progress.
The Convention then adjourned until Saturday at 10.30 a. m.

Saturday, March 16, 1912, 9.30 a. m.

Meeting of the Section on Building Blocks and Cement
Products, President Richard L. Humphrey in the chair.

The meeting discussed the materials, methods of manufacture
and tests of cement drain tile.

Digitized by

Google

SuMMABY OP Proceedings^ Eighth Convention. 27

Tenth Session — Saturday, March 16, 1912, 10.30 a. m.

President Bichard L. Humphrey in the chair.
The paper by George Gibbs on "Some Notes on the Value
and Comparative Cost of Reinforced Concrete Telegraph Poles,"
was in the absence of the author presented by the President.

The President read in the absence of the author, Robert A.
Cummings, the paper on "The Making and Driving of Rein-
forced Concrete Piles Within Six Days."

The paper by W. J. Towne on "Concrete Fence Posts,"
was in the absence of the author read by title.

A paper on "Comparative Tests of the Strength of Concrete
in the Laboratory and in the Field," was presented by R. J. Wig.
The following papers were, in the absence of the authors,
read by title:

"Field Inspection and Tests of Materials for Reinforced

Concrete," by G. H. Bayles.
"Unit Cost of Reinforced Concrete for Industrial Build-
ings," by C. S. AUen.
"Notes on "the Deformation in the Webs of Rectangular
Concrete Beams," by H. C. Berry.
The President then presented a paper by Alfred D. Flinn
on "The Use of Cement for Protecting Steel Pipes Along the
New York Aqueduct."

Robert F. HavUk presented a paper on "Modem Methods of
Manufacturing Concrete Products," which was followed by a
discussion.

The meeting then adjourned until 8.00 p. m.

Eleventh Session — Saturday, March 16, 1912, 8.00 p. m.

President Richard L. Humphrey in the chair.

The paper on "The Use of Concrete in the Fourth Avenue
Subway, Brooklyn, N. Y.," was in the absence of the author,
Frederick C. Noble, presented by the President.

W. A. CoUings presented a paper on "Reinforced Concrete
in Agriculture."

Digitized by

Google

28 Summary of Proceedings, Eighth Convention.

The paper by S. B. Williamson on "The Handling of Concrete
in the Construction of the Panama Canal," was in the absence of
the author, presented by the President.

The following papers were then read and discussed:

"The Present Status of Iron Portland Cements," by P. H.

Bates.
"Iron Portland Cement," by Herman E. Brown; in the
absence of the author, presented by the President.
The Committee on Resolutions, Rudolph P. Miller, Chair-
man, presented the following resolutions, which were unanimously
adopted:

(a) Resolved J That a eommittoo bo appoint od by the Executive Board
to consider the form of all standard specifications or recommende<i practice
issued by this Association with a view to securing uniformity so far as prac-
ticable.

(6) Resolved, That a committee of five members of this Association, of
which one member shall be Chairman of the Committee on Specifications
and Methods of Tests for Concrete Materials, be appointed by the President
to plan a comprehensive and systematic investigation of the aggregates used
for concrete and to interest State Universities, Experiment Stations and
other laboratories in carrying out the same.

(c) Resolved, That the Executive Board be instructed to consider the
advisability of appointing a Committee to report on Standard Specifications
for Concrete Highway Bridges and (Culverts.

(d) Resolved, That the Committee on Nomenclature be instructed and
empowered to extend its work to include the standardization of the size of
drawings, the symbols used on same and the graphical representation of
details.

Resolved, That the Committee on Cement Products be instructed to
consider the suggestions and criticisms on building block specifications offered
at this Convention, to confer with the Committee on Reinforced Concrete
and Building Laws with a view to reconciling there commendations of
the two committees, and to report revised specifications to the next
convention.

(c) Resolved, That a report be submitted to the next Convention on
Standard Specifications for Concrete Fence Posts and that the Executive
Board consider the advisability of having this done by a sub-committee of
the Committee on Cement Products or by a separate committee.

(/) Resolved, That the thanks of this Association are hereby tendered
the officials and the representatives of the local engineering and concrete
interests for their hearty welcome, to the citizens of Kansas City for their
co-operation in making this, the Annual Convention, a notable success, and
to the guests of the Association for their assistance in this success by the
contribution of their interesting and valuable papers.

Digitized by

Google

Summary of Proceedings, Eighth CoNVENTioN. 2§

(g) Reaolvedf That the thanks of this Association are hereby tendered
to the members who have aided by the presentation of papers, to the several
committees whose efforts have added this meeting to the long series of suc-
cessful conventions, to the local and technical press whose recognition of
the work of this organization is gratefully acknowledged, and to its officers
but particularly to its President, Mr. Richard L. Humphrey, for his untiring
devotion to the interest and welfare of this Association.

The President thereupon declared the meeting adjourned,
sine die.

Digitized by

Google

Digitized by

Google

National Association of Cement Users.

PROCEEDINGS

OF THE

EIGHTH CONVENTION

This ABSOciation is not responsible, as a body, for the statements and
opinions advanced in its publications.

EUROPEAN PRACTICE IN CONCRETE CON-
STRUCTION.

Annual Address by the President,
Richard L. Humphrey.*

Two years ago, in speaking on the use of concrete in Europe,
I discussed in a general way the progress that was being made
and contrasted the conditions prevailing in various parts of
Europe. It was also pointed out that in the artistic treatment
of concrete the foreign engineer and architect undoubtedly showed
greater skill than was shown in this country, and therefore obtained
much more pleasing results. I further commented on the fact
that the development of the use of concrete in certain countries
was very much handicapped by restrictive building laws. It was
my good fortune to again visit Europe last year and to inspect
extensively various structures of concrete, covering the most
important work west of the Russian boundary. I shall use this
opportunity for enlarging upon my former address, pointing out the
development and essential points of difference at the present time
in reinforced concrete construction in this country and in Europe.

This visit to Europe and the inspection of concrete construc-
tion during the closing months of last year, was under more favor-
able conditions than on the occasion of my previous trip, in that
I was a guest generally of the concrete associations, whose officers
did all in their power to show me everything of interest. This
was particularly true of my visit to Austria, where, as the Presi-

* Consulting Engineer, Philadelphia, Pa.

(31)

Digitized by

Google

S^ Annual Address by the President.

dent of this and Honorary Member of the Austrian Concrete
Association, I was the guest of the latter during my entire stay
in that country, and was under the devoted personal guidance
of its Director, Karl Bitner. This gentleman, as you know, was
a delegate to our New York Convention and one of the speakers
at the banquet, and I wish again to express my heartiest thanks
and appreciation for the courtesies ext^^nded to me and the
unusual efforts which he made to render my visit a pleasant and
profitable one. It is certainly true that this opportunity of
inspecting the concrete buildings in Austria was a valuable one
for the reason that some of the best examples of concn^te con-
struction are to be found in that country.

The Austrian Concrete Association has always manifested
a great interest in the work of our Institute and showed a willing-
ness to co-operate with us in every possible way. The unusual
character of the program laid out and the rapidity with which
various pieces of work were inspected was, I think, intended as
a tribute to our American characteristics. I submit this inter-
esting program. (See opposite page.)

A feature that impressed me most favorably was that my
visits to the various buildings had been arranged for in advance
and upon scheduled time. On our arrival at the building we
were met by the architect or his representative, the builder, and
the engineer in charge; in many instances the plans of the struc-
ture were tacked up at some convenient point, and before the
building was inspected a representative who spoke English ex-
plained the particular points of interest in the structure. On
Thursday night I was the guest of the Austrian ('oncrete Institute,
the Austrian Association of Cement Manufacturers, Austrian
Society of Engineers and Architects and the Austrian Clay Products
Association. My here recorded acknowledgment of the signal
honor conferred upon me but inadequately expresses th(» (l(»pth
of my gratitude and the extent of my appreciation. The precis-
ion with which the progran was carried out, the completeness of
the details and the warm hospitality extended to me by all those
I met, has left its permanent record — one that I shall never forget.

Dm-ing my visit to p]ngland it wius my privilege to address
the Concrete Institute in London, on October 26, 1911, on the
subject of " Fireproofing** for which I was honored by the award
of the Institute medal.

Digitized by

Google

OSTERREICHISCHER BETONVEREIN
WIEN, IV/2, MOLLWALDPLATZ 4

TELEPHONE NT. 10.597 TELEPHONE Nr. 10.597

ZEITEINTEILUNG

anlasslich

Des Besuches des Prasidenten der National Association of Cement

Users, Herrn RICHARD L. HUMPHREY, Consulting Engineer,

Philadelphia.

Donnerstag, 21. September, 1911.

9 Uhr 30 Min. Abfahrt vom Hotel Bristol:

9 " 35 " Kamtnerhofbasar, I. Kamtnerstrasse

(Ed. Ast ACic);

10 " 20 " Wohn- iind Geschaftshaus, I. Weihburggasse 7
(Plachy & Co.);

10 " 30 " Wohn -und Geschaftahaus, I. Weihburggasse 9
(G. A. Wayss& Cie.);

10 " 40 " Wohn- und Geschaftshaus, I. Weihburggasse 10

(N. Rella & Xcffe) :

10 " 50 * Lazzenhof, I. Rotenturmstrasse, verlangeter Fleischmarkt

(Kontrollbalkenversuch)
(k. k. Oberbaurat Dr. Ing. Friti von Empergcr — Chefingenieur
Richard Wucakowski)

11 " 30 *' Wiener Urania, I. Aspernplatz

(G. A. Wayss&Cie.):

11 " 55 *' Dreilaufferhaus, I. Kohlmarkt, Ecke Herrengasse

(Pittel & Brausewettrr) ;

12 " 00 " Wiener Bankverein, I. Schottenring

(Ed. Ast A Cie. und N. Rella & Neflfe);

12 " 30 " Lunch im Rathauskeller:

Digitized by

Google

2 Uhr 00 Min. Autogarage, II. Rembrandtstrasse 29

(Jancsch und Schnell) ;

40 " Basaltoidpflaster (2 Jahre alt), XVI. Hasnerstrasee—
Richard Wagner-Platz

(Baaaltwprke Radebeulo) ;
55 " Kirchenbau, XVI. Herbststrasse 66

(MazEtnerA Cie.);

25 " Zentralpalast, VI. Mariahilferstrafise — Ecke Kaiser-
strafTe

(G. A. Waysa & Cie.) ;

00 '' Gewerbliche Fortbildungsschule, VI. Mollardgasse

(N. Rella ANeffe):

20 " Schokoladefabrik Stollwerck, V. Gaudenzdorfer GOrtel
43,45

(Pittol A Braiwewett^r) ;

30 " Schule, V. Margaretenstrasse 103

(N. RellaANeffe);

15 " Kronenbrotwerke, X. Siccardsburggasse 83

(Ed. Ast A Cie.) ;

Wahrend der Fahrt zu besichtigen:
Kabelblocklegung ftir die Telephonleitungen, Unterleitung
der Strafisenbahn, Stadtbahn als Untergnind- und
Hochbahn etc. ;

00 ** Besuch des k. k. Hof-Opemtheaters;
Souper.

Freitag, 22. September.

Fahrt nach Berndorf, Besichtigung der Kirche, Weiterfahrt
nach Weissenbach, Besichtigung der Kunststeinfabrik;
Adolf Baron Pittel

Lunch;

Weiterfahrt;

Besichtigung der Zementfabrik Achau;

Besichtigung der Betriebsanlagen der Wienerberger
Ziegelfabriks^ und Baugesellschaft (Keramitfabrik und
Seidlbalkenerzeugung), X. Triesterstrasse 100;

00 " Souper am Cobenzl, angeboten vom Osterreichischen
Betonverein, Verein Osteneichischer Zementfabrik an-
ten, Osterreichischen Ingenieur- und Architektenverein
und Osterreichischen Tonindustrievcrein (bei schlechtem
Wetter Souper im Ktlnstlerzimmer des Restauj^ants
Hopfner).

8 Uhr 00 Min

Digitized by

Google

«9

" 16

" 30

" 45

" 05

Samstaflp 23. September.

8 Uhr 00 Min. Abfahrt vom Hotel Bristol:

8 '' 30 " Versuchsplatz des Eisenbetonausschusses des OsteireichiB-
chen Ingenieur- und ArchitektenvereineSi Wien, XIX.
Muthgasse;

SchleuBenanlagen in Nussdorf ;

Waaserturm am Bahnhof HeiUgenstadt
(N. Rella A Neffe):

Zigarettenpapierfabrik Schnabl, XIX. HeiUgenstadt

(N. Rella & Neffe);

Brtlcke im Zuge der Rampengasse, XIX.
fH. RelUftCo.);

10 " 20 ** Wagenhalle der stjidtbchen Strassenbahnen, XVIII.
Wahringer Gtirtel
(N. Rella ft Neffe);

10 " 45 " Physikalisches Institut, IX. Wahringerstrasse — Waisen-

hausgasse
(Ed. .\pt<fcCie.);

11 " 30 " Sargfabrik, XIII. Matznergasse 8

(A. Porr):

11 " 50 " Technisches Museum ftir Industrie und Gewerbe, XIV.

Ecke Winckelmannstrasse und Linzerstrasse

(A. Porr);

12 '' 30 '' Basaltoidflaster, XIII. Sch5nbrunner Schloesstrasse

(Basaltwerke Radebeule);

Lunch im Parkhotel (Hietzing);

Landes-Heil- und Pflegeanstalt Steinhof ;

Kohlenturm und Koksseperationsanlage im Gaswerke
XXI. Leopoldau
(H. Rella A Co.):

4 " 30 " Siloanlage nebst Weichraum und Lagerraum in der
Malzfabrik Hauser & Sobotka, XXI. Stadlau

(H. Rella A Co.);

Sehenswtkrdigkeiten von Wien;
8 " 00 " Souper Venedig in Wien.

Eventuell: Sonntag, 24. September.

Unter Ftlhrung des Herm Ingenieur E. A. Westermann,
Chef der Firma Wayss, Westermann & Cie., Graz:

Fahrt tiber den Semmering nach Graz, Besichtigung des
Landeskrankenhauses Graz und der Briickenobjekte an
der Bahnstrecke Weiz-Birkfeld.

" 46

" 15

" 30

Digitized by

Google

Montag, 28. Sentember.

Fahrt nach Retznei, Besichtigung der Zexnentfabrik,
Weiterfahrt nach Adelsberg, Besichtigung der Adels-
berger Grotte;

Fahrt nach Triest, Besichtigung der Hangarbauten,
Fundierungen etc.

Dienstag, 26. September.

8 Uhr 00 Min. Excelsior Palace Hotel

(Wayas' Wefltcrmann & Cie.) ;

8 " 30 " Greinitz

(Maxorana A Cornel) ;

8 " 45 " Riunione Adriatica

(Aat. ACie.):

9 '* 00 " Hanger -Bauten

(WayM* Freitag & Mciaong);

10-12 '' Rundfahrt im Hafen;

Eventuell: Souper in Opcina;

Fahrt nach Miramare.

Digitized by

Google

Annual Address by the President.

The various countries of Europe are progressing in the use
of concrete, but many of the large cities are still handicapped by-
restrictive building laws; particularly is this true of London,
where only recently have the London County Council Regula-
tions permitted the erection of structures that might be termed
reinforced concrete. Most continental countries show far greater
skill in the application of concrete than is shown in England,
where the British conservatism has resulted in heavy structures
of very simple application. This is also true for the most part
of the various cities in Germany. In France, in Belgiimi, and
particularly in Austria a wider and less conservative grasp in the
use of this material has resulted in the erection of structures

'AHDi JIMMf^H'^C

i4*'/l'*"l# KMVffil'J

FrKFf'dQOFING,

FIG. 1. — THE CONCRETE INSTITUTE MEDAL.

which are not equaled anywhere in the world. Certainly at the
present time in Vienna I believe one may find the most extensive
use of concrete that is to be found either in this country or Europe.
The city of Hamburg is perhaps second only to Vienna in the
number of its reinforced concrete buildings, and these two cities
are the most progressive spots in Europe. I observed on this
trip that more concrete buildings were in evidence in the out-
skirts of London and other English and continental cities than on
my previous visit; in this country the same is true, probably
for the same reason, viz., that the laws governing the erection of
buildings are more liberal outside of than inside of the larger cities.
The development of the use of concrete is certainly much
greater in Europe than it was two years ago. I do not think,
however, that Europe as a whole shows as great a development

Digitized by

Google

34 Annual Address by the t^REsiDEKf.

as is to be found in this count^>^ In certain parts of Europe,
Austria, for example, is shown a greater knowledge and skill in
the use of this material. In England, especially in London,
where there were practically no concrete buildings to be foimd on
my visit two years ago, there are now to be found many struc-
tures of reinforced concrete; although it is true that these are
not within the limits of the authority of the London County
Council, which permits only reinforced concrete for floors with
masonry-bearing walls, and to a limited extent for columns. It is
probable, since a recent revision of the London County Building
regulations permits the use of reinforced concrete, that from
now on many structiu'es wall be erected of reinforced concrete.

The materials in Europe available for use in concrete are
still relatively more expensive than labor. As a result, the
designer, for purposes of economy, finds it desirable to so shape
his forms as to eliminate as much as possible all material which
is not required for structural or protective purposes. The effect
of this is to render the structure less massive and more pleasing
in appearance. The abundance of extremely cheap unskilled
labor and the presence of low-priced technically-trained labor is
one of the great advantages in the erection of concrete struc-
tures in Europe. This is particularly true as to the foremen
and labor bosses who, in many cases, especially in Germany, are
technically-trained men, which is of course unusual in this country.

Another feature which tends to increased efficiency in the
erection of concrete structures is the fact that there are govern-
mental regulations which apply with sufficient rigidity to terri-
tory outside of the large cities. In large cities the regulations are
necessarily much more rigid. There is a wholesome respect for
the law throughout Europe — ^which is lacking in this country —
with the result that each person concerned conscientiously en-
deavors to erect the structure in full accordance with the building
regulations. In some countries, especially Germany, a contractor
who is found guilty of dishonest practices loses caste and becomes
discredited, which is, after all, the most effective way of pre-
venting the construction of dishonest structures. I recall par-
ticularly a case in Stuttgart where what might properly be called
a "Quantity" engineer, having assumed responsibility for the
erection of the building, was arrested and sentenced to several
years imprisonment by reason of the collapse of the structure of

Digitized by

Google

Annual Address by the President. 35

which he had assumed the responsibility. Another effective way
of dealing with this subject is practiced in France, especially in
Paris, where the building department acts merely as a custodian
of plans; the law placing the responsibility for the structure on
the architect, contractor and owner. Under the law, these three
parties are held to be responsible for the collapse of the structure
until their innocence is established. The use of cheap labor,
especially women, to carry mortar and concrete in small tubs on
the head, seems to be the usual method but hardly an effective
one for handling concrete. In the erection of but two buildings
did I see used the elevating machine, so commonly used in this
country, for handling concrete.

The presence of a large number of women laborers on con-
crete structures was always a source of interest to me as was
also the pittance that these women were paid for a day's work.
A recent lecturer in New York stated that in African hunting
expeditions the camp followers received about sixty cents a month.
While these women do not receive so little, yet when you con-
sider that in the one case the men were furnished their food,
while in the other, the women had to supply it themselves, the
wage paid the women (in some countries equal to 16 cents and as
low as 12J cents a day) is extraordinarily low and you can appre-
ciate why elevating and conveying machinery is doubtless more
expensive than labor. The almost imiversal limit of about 5
stories or 22 metres in the height of all structures renders ele-
vating and conveying machinery relatively unimportant. When,
however, speed in the erection of concrete structures becomes
important, Europe will be obliged to resort to mechanical means
for elevating and conveying concrete, as the method of carrying
concrete in tubs on the heads of the laborers is too slow for proper
continuous placing.

Fig. 2 is a view of the memorial church at Berndorf, the
industrial village of the Krupp works, just outside of Vienna,
in which will be seen a nimiber of the women laborers who are
engaged in carrying mortar and concrete in the manner above
described. These women, in spite of their skirts, are able to climb
ladders with almost the same speed as men.

The use of round timbers instead of sawed, as studs for forms
and scaffolding, is quite general. Where a splice is necessar>'^,
the two parts are tied together with rope or chain. They do

Digitized by

Google

AN^UAL^ApbRESS Bt I'&E t^RESlfiENT^.

not use nails, therefore, these timbers can be used over and over
again and last for such a length of time as to make the cost rela-
tively very low. Another interesting construction detail is the
method of splicing uprights through the medium of two iron
rectangular iron bands siurounding the uprights, which are clamped
tightly together by wedges between the inside of the band and the
face of one of the uprights, which is a very simple and readily
adjustable device. This particular device was used on the props
in the construction of the Commercial Museum in Vienna.

The development in the use of reinforced concrete telephone

na. 2. — DOME MEMORIAL CHURCH IN BERNDORF, AUSTRIA.

and telegraph poles is even greater than it was on my previous
visit and the experience gained has brought forth many orna-
mental and efficient poles; the tendency for purposes of economy
is towards a hollow pole, this being of greater necessity in Europe
than in America; and in my judgment the cost of the concrete
pole must be materially reduced, to effectually compete with the
wooden pole in this country. The view shown (Fig. 3) of the
centrifically molded circular poles in Bad Kosen indicates the
uniformity and symmetrical shape of this pole. This circular
form, however, is not necessary, and there are many hollow poles
of ornamental character of square or octagonal design, notably

Digitized by

Google

Annual Address by the President.

four in Dresden, in the Exposition of Hygiene. The ornamental
poles on the Augusta Bridge in Dresden further exemplify the
beauty of this type of pole, the hollow interior affording an ideal
place for running the electrical wires.

The method on making the hollow reinforced concrete pole

FIG. 3. CENTRIFUGALLY MOLDED CIRCULAR TOWN ELECTRIC LIGHT POLES,

IN BAD KOSEN, GERMANY.

consists in placing the mold filled with concrete in a machine
rotated at the rate of 600 revolutions per minute. The effect of*
this high speed is to force the concrete against the walls of the
mold by centrificial action, gradually compacting the concrete
and forming a hollow space in the center of the pole in which

Digitized by

Google

38 Annual Address by the President.

the excess water and laitance gathers producing an exterior sur-
face of a hard and uniform texture which greatly adds to the
appearance of the pole. The molds are generally kept on the
pole until properly hardened.

The poles erected in connection with the Danish railways in
Copenhagen, Denmark, are hollow, 36 ft. high, and were molded
by the hand process and the tests showed them to be very stiff and
capable of resisting high loads.

It has been found in Europe, especially where lumber is
becoming scarce, that reinforced concrete poles are much more
efficient and less costly than wooden poles; that the maintenance
of the line is less expensive and the permanent life of the pole
much greater. It appears to me to be evident that in order to
effect the desired economy in the cost of manufacture of this
class of concrete products, it is necessary to turn out a great
many each day; inasmuch as they are coming into general use
in Europe, there is a constant decrease in the cost of manufacture.

The poles illustrated in Fig. 4 are those at the plant of R.
Wolle in Leipsic, Germany. There seems to be a general use for
concrete poles for carrying high-tension lines, especially where
the pole must be of considerable height. It is claimed that the
cost of maintenance of such lines is very much less than for
wooden poles. This probably accounts for their popularity.

The continued use of reinforced concrete poles in this country
leads to the belief that as the number of poles and skill in making
them increases they will become more serious competitors of the
wooden pole and will in time replace the other forms of telephone,
telegraph and electric transmission poles. They can be molded
to suit the particular conditions of almost any height and can be
so anchored in the ground as to enable them to maintain a rigid
position in almost all soils. The high tension transmission line
poles of reinforced concrete, used in connection with the Penn-
sylvania Railroad tunnels,* which were erected on a mattress in
the marshy land of the approaches on the New Jersey side, are an
illustration of the superior excellence of this type of pole.

Another matter which was of considerable interest to me
was the form of chimney developed by Captain F. Mohl in Copen-
hagen. This consists of a four-leaf-clover section at the base

♦See rroc, N.A.C.U., Vol. VIII, p. 769,

Digitized by

Google

Annual Address by the President.

which gradually merges mto a circular at the top. A num-
ber of these chimneys have been erected, some of considerable
height, and it is said that they are more economical and much
more suitable than the ordinary circular stack.

The reinforced concrete barges shown in the illustration were
photographed (Fig. 5) at the place of manufacture in Livomo,
Italy. These particular barges were used for handling coal and
seemed to be in every way thoroughly satisfactory. When the
Italian government first used reinforced concrete for armor plate,

FIG. 4. — TYPES OF REINFORCED CONCRETE POLES AT PLANT OF R. WOLLE,

LEIPSIC, GERMANY.

there was much amusement manifested in this country and the
average person believed that the weight of this material would
sink the boat. Its use, therefore, to form the entire hull of a
boat would seem even more quixotic. Barges and other vessels,
especially battleships, are made entirely of steel, which is heavier
than concrete; and when you consider that the floating of the
vessel is a question of buoyancy depending on the lightness of
the material and character of the air-tight compartments, it is
evident, I think, that any material properly designed will b^

Digitized by

Google

Annual Address by the President.

suflSciently buoyant for practical purposes. In the case of the
armament of concrete, which is much lighter than steel, the thick-
ness can be much greater and the toughness of the former material
renders it a better protection. I believe with the development
of the art of reinforced concrete boat construction, these vessels
will in the future come into general use and will prove most ser-
viceable and economical, both as to first cost, maintenance and
durability.

It is not so very many years ago that the concrete barge was
a novelty and regarded by many as a freak application of cement.
However, a few years' trial of these boats has resulted in striking
economy and in many places I found concrete barges being used.

FIG. 5. — REINFORCED CONCRETE BARGE FACTORY AT LIVORNO, ITALY.

especially of the canal boat type, for handling coal and other
materials. It has been found that the durability and serviceability
of these boats render their ultimate cost very much less than
boats made of any other material. There have been a number
of such boats used in this country, notably in connection with
the construction of the Panama Canal, and in my judgment
there will be an even greater use of them in the future.

The constant study of the reinforced concrete railroad tie
(with somewhat disappointing results at the present time) shows
a desire for a tie of this type. In parts of Europe where steel or
wooden ties are readily obtained at reasonable cost, the concrete
tie does not make much headway. In other parts where ties of

Digitized by

Google

Annual Address by the President.

timber and steel are expensive, for instance in Italy, where the
steel and wood tie is at least as expensive as the concrete, there
has been considerable development and the officials informed me
that more than 300,000 were in use in the Italian railways and
that there were contracts for upwards of a million. The tie,
however, has not a very great life when used in main line ser-
vice, where it is subjected to the frequent passage of heavy loco-
motives. In what are termed secondary lines and sidings, ties

FIG. 6. — ^REINFORCED CONCRETE TIES IN MAIN LINE ITALIAN RAILWAY AT
PORTO NACCIO, ITALY.

of reinforced concrete are reasonably effective and in Italy have
as much as six or seven years of life.

I inspected some railroad ties just outside of Rome and
foimd that these ties (see Fig. 6), which had been in service for
about two years, were not wearing very well. A number of ties
having crushed just inside the rail.

The method of fastening the tie consists of the use of a
wooden block, cylindrical in shape, which is driven into the hole
molded in the tie and au ordinary wood screw which fastens

Digitized by

Google

Annual Address by the President.

the rail to the tie. When the threads in the block are worn by
use, the block is replaced.

Fig. 7 is an illustration of the section of track of the London
and Southeastern Railway at Knockholt station in which rein-
forced concrete ties have been used. These ties, I think, are not
of as good design as those used in the Italian railways. They
have been in service about two years. A number of them have
failed in the manner shown in the illustration (Fig. 8) by the
concrete crushing just inside the rail. It is my opinion that the
difficulty in reinforced concrete ties is in a lack of proper analysis

FIG. 7. — REINFORCED CONCRETE TIES IN SECTION OF LONDON AND SOUTHEASTERN
RAILWAY AT KNOCKHOLT STATION.

of the stresses, and that a tie could be so designed as to properly
care for these stresses and thus prevent the breaking down of the
tie in the manner just referred to. With this point cared for the
life of the tie would be greatly prolonged.

It was the universal opinion of track men that through the
use of reinforced concrete ties the cost of maintenance could be
materially reduced and the alignment of the track much more
readily maintained. It is, however, on curves that the ties are
least effective and their Ufe very brief. Another objection seems
to be that the use of the concrete tie usuallj^' results in a rigiditjr

Digitized by

Google

Annual Address by the President.

of roadbed which is extremely undesirable from an operating
point of view, with the result that many devices have been tried
with a view to introducing some elastic medium which will absorb
the shock of passing locomotives.

At the Exposition in Turin were shown a numl^er of rein-
forced concrete ties with wood blocks, fiber cushions, and other
similar elastic shock al)sorbers, placed in the tie with a view to
increasing its life. In most cases the design of the tie seemed

FIG. 8. — MANNER OF FAILURE OF REINFORCED CONCRETE TIES IN LONDON AND
SOUTHEASTERN RAILWAY.

to be at fault; many of them had been developed by mechanics
not skilled in structural designing, with the result that the rein-
forcement was not properly placed with regard to the stresses,
especially those of impact; and it appears to me that the con-
crete tie problem can only be solved with a due consideration
for these stresses.

Every one of the railroad officials who has had to do with
concrete ties in Europe feels sure that a tie will be developed
which will overcome the objections above indicated and redyc^

Digitized by

Google

44 Annual Address by the President.

the initial cost to a point where its life and reduced cost of main-
tenance will make it the cheapest railroad tie.

The objection in this country to the reinforced concrete tie,
namely, its rigidity under moving loads and the consequent objec-
tionable hammering to the locomotive, can be cared for by the
introduction of an elastic cushion in the shape of fiber or wood.
I believe the concrete tie will be the tie of the future and its first
cost will prove of minor consideration when the reduced cost of
maintenance and durability is considered.

Another interesting development in the use of concrete is in
connection with sewers or conduits of circular or elliptical form
in which the walls are made up of segments of concrete, which
after being placed in position are grouted or cemented together,
forming a solid ring. There were a number of cases where unusual
economy had been effected through the use of these sewers. In
some of them the segments were grooved along the axis and
around the circumference in which the reinforcement was placed.

The use of reinforced concrete for the lining of sewers and
tunnels, in my opinion, is a very important application. The
possibility of molding these blocks and placing them in position,
and filling the space between the roof and the ring with con-
crete forms a very simple and effective method of tunnel lining,
the application of which is cheaper than brick and also cheaper
than concrete construction where tight forms must be constructed
and maintained in position until the concrete has properly hard-
ened. In the segmental method, with the completion of the ring,
the concrete backing may be readily placed in position and but
little shoring will be necessary until the concrete has set.

Another extremely interesting matter was the use of rein-
forced concrete pipe of varying lengths which was laid as shown
in Fig. 9 to conform to the general contour of the ground, the
connections and adjustment being effected by means of loose
sleeves slipped aroimd the joints. After the pipes and sleeves
are in position the spaces between the sleeves are grouted solidly
to the pipe This appUcation is in advance of anything we are
doing in this country. Sections of reinforced concrete pipe laid
in this manner, in lengths of 12 feet or more, would have great
possibilities for use in pressure lines. The pipe in the illustration
were made in a machine not unlike a lathe. The reinforcement

Digitized by

Google

Annual Address by the President.

Was placed around a core rotating horizontally and a very stiff
mortar thrown against it. After the mortar had been built out
to the requisite thickness, the pipe was wrapped with cotton
bands four or five inches in width. The pipe was then removed
and the wrapping kept wet until the mortar had set, when the

1 %

>.2 "^i

A.4. .'

FIG. 9.-

-REINFORCED CONCRETE PIPE OF VARYING LENGTHS IN USE IN
SWITZERLAND.

bands were pulled off. This method of construction naturally
requires a great deal of labor and even if considered desirable
would be entirely too expensive for use in this country.

Fig. 10 shows a cement products yard in Vienna and par-
ticularly illustrates the concrete ducts which are being used in

Digitized by

Google

Annual Address by the President.

that city. These ducts have longitudinal recesses in which rein*
forcement is placed and the remaining space filled with cement
mortar; this stiffens the ducts and holds them in .position. It
seemed to me there was an excellent opportunity for this type
of construction in this country. When properly cemented in
position, the ducts formed were in every way desirable and I am
told that the cost is much less than that of terra cotta or other
material.

During my last visit in Vienna I had occasion to refer to the
use of concrete pavements and. I find that this type is coming

FIG. 10. — CEMENT PRODUCTS PI/ANT OF PITTEL AND BRAUSEWETTER AND
E. GAERTNER IN VIENNA, AUSTRIA.

into gradual use. The pavement inspected on my previous visit
and reported to you two years ago had been laid four or five
years and during my last visit I found these concrete roadways
still in excellent condition and that they had not been repaired.
The city of Vienna was engaged in laying considerable yardage
of these pavements, especially around the Royal Palace. This
pavement was of great width and was entirely of concrete. I am
absolutely convinced of the durability of these pavements and
repeat what I have stated in my previous address, that I believe
they will come into general use.

Digitized by

Google

Annual Address by the President.

The tank of reinforced concrete in use at the coal mines,
in Rotherham, England, to remove the coal dust from the water
coming from the coal washery is of considerable interest. This
water is pumped into the tank, the coal dust is removed and
the clean water returned to the washery. This coal dust is used
in briquetting coal. Steel tanks are not available for this purpose
because the sulphur in the coal would so corrode the steel as to
make its life extremely short. The concrete tank has proved
highly satisfactory not only in its resistance to the action of
sulphur but in its water-tightness.

I commented on the artistic use of reinforced concrete in
bridge construction in my previous address, but the matter is
so striking that I cannot help referring to it again. The innum-

FIG. 11. — BIEMORIAL REINFOKCED CONCRETE BRIDGE CONNECTING HISTORICAL
AND ART EXPOSITIONS OVER TIBER, IN ROME, ITALY.

erable bridges throughout Europe, carefully designed, carrying
railroad as well as ordinary highway traffic, are monuments to
the ability of the European engineer. The designs are for the
most part graceful and show a wide diversity in artistic treatment.

A structure of great beauty is the Memorial Bridge (Fig. 11)
in Rome connecting the Historical and Art Expositions, located on
opposite sides of the Tiber. This bridge was built by the Henne-
bique Construction Company and is an example of European
engineering skill. By reason of floods, it was necessary to con-
struct the centering on which this bridge was built, of reinforced
concrete — ^which is unusual and the first application of the use
of reinforced concrete for centering that has come to my at-
tention.

The bridge has a span of 100 meters and is built in imitation

Digitized by

Google

48 Annual Address by the tRBSil)ENt.

of the native Travertine stone; the beauty of this is fully as
great as the artificial Travertine stone to be found in the New
York station of the Pennsylvania Railroad. This structure was
built by the Italian government, and many severe tests were
applied with but extremely slight deflections.

The manner in which this bridge was tested is also of interest.
A commission was appointed to conduct the tests for the govern-
ment. These tests consisted in moving heavy steam rollers and
marching soldiers in a solid mass over the bridge; particularly
interesting was the loading of the bridge solidly with soldiers
and studying the effect of their cadenced step on the structure.

FIG. 12. — REINFORCED CONCRETE TRUSS 113.8 FT. SPAN USED IN MAIN RAILr

ROAD STATION IN LEIPSIC, ERECTED FOR TESTING IN COSSEBAUDE,

NEAR DRESDEN, GERMANY.

I think it may be stated without fear of contradiction that
reinforced concrete structures properly designed are less affected
by vibratory and cadenced loads than any other structure. It
is particularly noticeable that in structures of identical design
as to carrying capacity, those of steel show more movement than
those of reinforced concrete.

A most interesting structure is the truss (Fig. 12) erected
initially by the firm of Dyckerhoflf and Widman, at their plant in
Dresden. It was erected for testing the actual strength and justify-

Digitized by

Google

Annual ADUREgd by fHE President. 4§

ing the use of a truss of this tjrpe in the main railroad station in
Leipsic. The arch was loaded in various ways and the deflections
observed, and the tests were so satisfactory that it was successfully
used in the structure. It is unusual in this coimtry to go to the
expense of erecting a structure of this kind and applying tests in
order to satisfy the officials that it is amply safe; It would, how-
ever, be an effective means of preventing failures.

A matter that impressed me was the development of struc-
tures, many of them articulated, composed of separately molded
parts, and the views herein presented exemplify some of the more
remarkable of these structures. The practice of casting mem-
bers and then pinning them together and encasing the connection
in concrete is a development that seems to meet with favor, and
while this may not appeal to many of the conservative engineers
of this country, it seems to me that when this pin connection is
of the same design as pin connected steel members, there is no
reason why this type of structure, when properly designed, should
not be more serviceable by reason of the concrete covering which
makes it more durable than a structure of steel. The concrete
trestle used in connection with the mine at Floreiffe, Belgium,
the lower portion of this is used for the county roadway and the
upper portion for the handling of cars to the "tipple" from the
coal mine, is an illustration of a remarkable development of
this principle. I was informed that the parts of this structure
were separately molded and afterward erected in place in a very
short space of time and that the cost was considerably less than
a similar structure of steel or of timber. There is a tendency
towards systems requiring separately molded members, and it
seems to me that such systems afford an opportunity for economy.
The weak points in such structures are the joints, and with the
same attention to these connections that is given to steel struc-
tures there is no reason why the joint in a concrete structure
should not be stronger than the weakest part of the separate
members.

A development of the Visintini system was to me somewhat
surprising, since this system is very little used in this country;
the early attempts to introduce it were unsuccessful, chiefly
because of its cost as compared with that of other systems of
reinforced concrete construction. In Europe it has a wide appli-

Digitized by

Google

50 Annual Address by the President.

cation for buildings and bridges. Fig. 13 shows the bridge
being erected in Copenhagen, Denmark, with girders of system
Visintini. This bridge, erected in connection with the new sta-
tion that the State Railways were building, carries a street with
two Hues of electric trams over the railroad. The ornamental
portion is entirely of molded concrete which was not subsequently
treated. Where labor is inexpensive the cost of forms is not so
important as the cost of material. In this country, where condi-
tions are just the reverse, this system proves uneconomical. The
bridge consists of deep girders spanned by beams all of the Visin-
tini system. The latter spaced solidly so as to form a slab, under
the railroad tracks and 20 inches apart between the tracks. I

FIG. 13. — HUmWAY BRIDCIK OF REINFORCED CONC^RETE OVER STATE RAILWAY
TRACKS, COPENHAGEN, DENMARK.

think this bridge might be said to be a type of unit construction.
Certainly the bridge, which was nearly completed, presented an
extremely beautiful, ornamental appearance and far more desir-
able than a similar structure of steel. I believe that this method
of construction, which is in a measure illustrated in the flat slab
construction in use in railroad bridges in this country, possesses
great possibilities in the matter of appearance and speed of erec-
tion. All the beams of the system could be molded and the
structure then erected continuously, replacing perhaps an old
structure, without interfering with the traffic.

An interesting type of girder construction is shown in Fig. 14,
a highway bridge at Desna, Austria, which consists of a series of
Visintini girders supported on piers of rather unusual construction.

Digitized by

Google

Annual Address by the President.

The application of this system for arch ribs of a building is illus-
trated in (Fig. 15) the ceiling of a church erected in Aussig,
Austria, and which I think is also an application of unit system.
The lower flange of the rib of the arch is extended so as to support
the Visintini beams which form the ceiling. Another type of
structure which shows a trend in Europe that conservative struc-
tural engineers in this country might call dangerous is illustrated
Fig. 16. This is a trussed bridge practically a development of
the Visintini system, which forms a central span, with Visintini
girders and beams used in the approaches.

FIG. 14. — REINFORCED CONCRETE HIGHWAY BRIDGE, DESNA, AUSTRIA.

The photograph, Fig. 17, shows the coal bunkers of the
extensive plant of the city gas works in Vienna. The one on
the left side has most of the forms removed and illustrates the
generally pleasing character of the structure, which is notable in
view of the fact that it must be massive, in order to carry a large
quantity of coal. The attempt to render this structure pleasing
and ornamental could well be emulated by our American designers.
The exterior surface is dressed with pneumatic tools and a color
effect has been obtained which is not unlike that attained in the
construction of the Connecticut Avenue Bridge in Washington.
D. C.

Digitized by

Google

Annual Address bv thE President*

no. 15. — PORTION REINFORCED CONCRETE ROOF OF CHURCH, AUSSIG, AUSTRIA.

FIG. 16. — REINFORCED CONCRETE TRUSS HIGHWAY BRIDGE NEAR VIENNA,

AUSTRIA.

Digitized by

Google

Annual Address by the PREsroENT.

We have, of course, many water towers in this country, but
the thing that most impresses me in connection with those in
European countries, is the thinness of the tank walls and the
fact that they are constructed without the aid of waterproofing;
the density of the concrete and the manner of reinforcing is suf-
ficient to so distribute the cracks as to render them water-tight.

In connection with buildings abroad it would seem that more
attention is given to the exterior and interior finish, even in
factory buildings, than is given in this country; the skill used in

FIG. 17. — REINFORCED CONCRETE COAL BUNKERS, CITY GAS WORKS, VIENNA.

secimng a pleasing finish, even considering the low-priced labor
of Europe, is very rarely more expensive than the rough finish
commonly used here. The tendency toward flat slabs and the
elimination of as many beams as possible shows, I believe, an
unmistakable turn, which is reflected in this country in the recent
development of the flat slab type of construction. In many
places the elimination of beams results in paneled effects through
the use of girders between the columns and mortising and mold-
ing the connection between the slab and girder in such a way
as to produce pleasing ornamental effects. There is also a ten^

Digitized by

Google

Annual Address by the President.

dency to panel the slab itself, which relieves its flatness and adds
no little to the beauty of the interior finish. The use of much
higher ceilings than are commonly used in this country renders
flat slabs unnecessary for the distribution of light. The effect
of the paneling is somewhat that of the Roman barrel arch, and
this latter type of construction may be seen in the ceiling decora-
tion of many European structures.

An excellent exterior finish having the appearance of gray
stone was to be seen in the C'igarette Paper Factory in Vienna.
This structure had been erected many years and the weathering
had been so uniform as not to in any way mar its beauty. This
building was illustrated in my address describing my trip of two

iiillri

111.
liiUjiiiliii fliirrn

111

iii
nil

FIG. 18. — TRADE SCHOOL OF REINFORCED CONCRETE, VIENNA, AUSTRIA.

years ago and is again referred to because after an interval of two
years the exterior finish of the building remains unchanged. The
entire ornamentation is of concrete and serves as an excellent
illustration of the artistic possibilities of this material.

The Trade School in Vienna (Fig. 18) is also an excellent
example of the artistic treatment of concrete. The paneling and
ornamental work are most excellent in character, and for this
reason do not call forth the criticism which once was so rife in
this country, where the crude structures which we erected left
much to be desired from the aesthetic point of view. It is fre-
quently the practice in Europe, especially in Vienna, to cast
monolithic walls and tool them afterwards, in a manner to pro-

Digitized by

Google

Annual Address by the President.

duce the effect of stone. This, however, does not appeal to me
so strongly as the structure in which there is no attempt to
imitate stone, but where the material is used to stand for what
it is, producing pleasing ornamentation and a surface uniform
in color and texture and free from those stains and cracks which
are frequently seen in structures of concrete in this country.

FIG. 19. — CITY HALL OF REINFORCED CONCRETE, NEAR BADEN, AUSTRIA.

Fig. 19 is a view of the City Hall at Weikersdorf near Baden,
Austria. The entire wall, with ornamentation, is of concrete
without any attempt to imitate other material, which I think
illustrates that this is the proper way in which to use concrete
and that when so used the results are much more effective than
where it is used in imitation of other materials.

A favorite form of exterior decoration is to apply phvster

Digitized by

Google

Annual Address by the PREsroENT.

to the rough concrete and mold it in a manner to produce artistic
effects.

The photograph of the Villa Figari in Genoa, Italy, shown
in Fig. 20, is an illustration of the splendid possibilities in the
artistic use of reinforced concrete. This graceful, beautiful struc-
ture needs no comment. The erection of this Villa involved a
number of interesting problems iVi design which I think were met

FIG. 20. — VILLA FIGARI, GENOA, ITALY.

more successfully through the use of this material than would be
possible through the use of any other. There are a number of
plants in this country engaged in making ornamental concrete
products of a very high order, and fully equal to some of the
best work done in Europe, but the American designer does not at
the present time seem to appreciate the adaptability of concrete
for use in the ornamental structural portions of buildings; it
seems to me that development along the lines illustrated in the
Villa Figari opens a wonderfully promising field.

Digitized by

Google

Annual Address by the President.

As the architect and designing engineer more fully realize
the architectural possibilities of concrete and apply them in build-
ing construction with a due appreciation for the aesthetic, there
will result such graceful structures as will entirely meet the
criticisms and objections to the many cumbersome, ungraceful
and unattractive structures which are being erected at the
present time.

The desire to render even ordinary structures beautiful is

FIG. 21. — REINFORCED CONCRETE CAR BARNS OF CITY OF VIENNA, AUSTRIA.

particularly observable in Europe and it is, therefore, unusual
to see even a mill building in which no attempt has been made
to give it a proper finish. The eye for the beautiful is a matter
of education and growth and I presume that in time we will
develop similar tastes so that our mill buildings will be given
artistic finishes. It should be borne in mind, however, that the
cost of this work in Europe is considerably less than in this
country, by reason of the cheap labor and because of the greater
quantity that is done; they have acquired more skill and there

Digitized by

Google

58 Annual Address by the President.

are a great many more skilled men capable of doing this class
of work.

The car barns of the city of Vienna, shown in the illustration
(Fig. 21), is another development in the use of concrete which
I think will become general in the future. This structure is built
entirely of concrete, including the roof which has not been water-
proofed. I understand that the cost of this structure was mate-
rially less than the cost of other types of construction. The
span and lines of the girder was a matter of interest and is char-
acteristic of the skill of the Austrian engineer.

A building of reinforced concrete which was brought to my
attention during my visit two years ago, and which was then
under construction, is the Urania Building in Vienna, which is
devoted to the development of popular musical education. It
has a rather difficult problem in acoustics which is of moment
to those interested in concrete. The main auditorium has a floor
and ceiling of reinforced concrete of considerable thickness;
above and below this auditorium are smaller halls for musical
entertainment. The large auditorium contains an organ, and
there is no connection between it and the halls above and below,
yet the Director states that when there is a concert in progress
in the large auditorium it is impossible to hold a concert in either
of the other auditoriums, because the sound of the music in the
large hall is so pronoimced as to seriously interfere with the per-
formance in the other halls. This subject has received a great
deal of consideration in Vienna and methods are now in progress
to eliminate, if possible, by some insulating medium, the trans-
mission of sound.

Another matter of interest was the work which the Austrian
Association is doing in investigations and tests, not unlike those
of our own Committee on Reinforced Concrete, except that the
value of the work accomplished is greater because the work is
much more extensive. They have conducted a comprehensive
series of tests of columns and beams, the cost of the work being
defrayed by the Austrian cement manufacturers, and has been
under the auspices of representatives of the government, the
Society of Engineers and Architects, the Austrian Concrete Asso-
ciation and the Austrian Association of Cement Manufacturers.
The principal feature of the tests which were being conducted at

Digitized by

Google

Annual Address by the President.

the time of my visit was a study of the effect of the weight of
wall in restraining the ends of beams; a number of different
methods were being tried and it had been found that the stiff-

PIG. 22. — A RESTRAINED REINFORCED CONCRETE BEAM AFTER TESTING.

FIG. 23. — METHOD OF APPLYING LOAD IN TESTING REINFORCED CONCRETE

BEAMS.

noss of the beam was materially increased through the weight
of the wall.

Fig. 22 illustrates one of these beams after test. In some

Digitized by

Google

60 Annual Address by the President.

cases the beam rested on a pier, and in others it extended over;
in still others it was imbedded in concrete, or laid up in a brick
wall so as to approximate practical conditions.

Fig. 23 shows the manner of loading and observing the deflec-
tions. The method of applying the load is perhaps of interest.
The load to be apphed was carried by hydraulic jacks, the lower-
ing of which brought the weight of this super-imposed load on
the beam.

One thing that was apparent on my recent trip, and which
must be a matter of great gratification to every member, is the
high regard in which this Association and its work are held in
Europe. It is taken generally as a model, and in many cases
the character of the work done here and the value of its pro-
ceedings and discussions are so appreciated as to result in a
number of Europeans becoming members in order to secure its
publications.

There is also a cordial feeling of cooperative good-will between
the various concrete organizations of Europe and our Association,
which I hope may be fostered and that it may be possible through
the interchange of delegates, papers, and in other ways, to extend
this cordiality so as to obtain the advantages of the development
in the art of concrete construction in various parts of the world.

This Association stands for education in the development
of the proper use of concrete, and certainly it should be a part
of its policy to encourage international co-operation, to the end
that the development in this country may proceed with a full
knowledge of what is being done by our foreign competitors.

Digitized by

Google

REPORT OF THE COMMITTEE ON REINFORCED
CONCRETE AND BUILDING LAWS.

In accordance with the program of tests on completed
stnictures presented to the Association at the last convention,
four tests have been made, two of them on standard forms of
reinforced concrete floors in which the concrete slab is supported
by one or more intermediate beams which in turn are framed
into or supported by girders carrying the load to the columns.
Of the other two tests one is on the girderless or flat slab floor,
and the other on a combination of tile and concrete floor, in which
the reinforcement is placed in two directions.

Making arrangements for the tests* and computing the
results have taken up all the available time of the Conmiittee;
the test data is therefore presented to the Association without
any attempt at generalization.

The tests on the Wenalden and the Turner-Carter buildings
were made under the direct supervision of A. N. Talbot; the
other tests were made by F. J. Trelease of the Research Depart-
ment of the Corrugated Bar Company, with the assistance of
W. A. Slater of the Illinois Engineering Experiment Station.

Part I. Tests op Two Reinforced Concrete Buildings
OF the Beam and Girder Type.

Preliminary, — These tests were undertaken for the purpose
of obtaining information on the action of the composite structure
of concrete and steel under load in a reinforced concrete building
constructed under the usual conditions of work. Many tests
have been made of separate reinforced concrete members, but
little attention has been given to the measurement of stresses and
deformations in the completed building and to the determination
of their actual amount and distribution and of the effect of one
part of the structure upon another. Load-deflection tests are
common and are of value in judging of the quality of workman-
ship and in giving confidence in the structure, but they throw
littie light on the stresses developed in the different parts or

(61)

Digitized by

Google

62 Report of Committee on Reinforced Concrete.

upon their distribution. A variety of views have been advanced
on the relation between the bending moment at a section at the
support and that at the middle of the beam, on the amount of
arch action which may be developed in the structure, on the dis-
tribution of the stresses across a floor slab acting as the flange
of a T-beam,»on the restraint of girders and beams, etc. These
tests were undertaken in an effort to obtain some information

FIG. 1. — INSTRUMENTS AND TOOLS.

on such matters as well as to find something of the general action
of reinforced concrete structures as a whole.

The general method of test followed the plan outlined by
Arthur R. Lord in the paper, A Test of a Flat Slab Floor in a
Reinforced Concrete Bxdlding* presented at the New York Con-
vention. Holes were cut in the concrete until the reinforcing
bars were bared. Gauge holes were then drilled in these bars,
at distances apart to give the proper gauge lengths. Where
measurements of deformation of the concrete were desired, holes
were cut in the concrete and a metal plug inserted in which the

♦ See Proceeding, Vol. VII, p. 156.

Digitized by

Google

Report op Committee on Reinforced Concrete. 63

gauge holes were later drilled. These gauge lines were selected
in places where it was thought that critical stresses would be
determined. In some places for one reason or another the rein-
forcing bars were inaccessible and it was impracticable to obtain
measurements to give information which would have been of
interest. In some cases a series of gauge lines were used to de-

/>-o/T» C rt> //."

Central ho/^

Eccentric hole

Finishing
tooL

FIG. 2. — SHOWING EXTEN80METER AND GAUGE HOLES.

termine the change of stress or distribution from one point to
another as at the end of a restrained beam and across the floor
slab between beams.

The measurements were made by means of Berry exten-
someters of the form developed at the University of Illinois. The
extensometer is shown at the bottom of Fig. 1. The instrument

Digitized by

Google

64 Repor* of Committers on Reinforc*sd CoNCRETti.

reads to Wajf in. and is estimated to TJi^nis in. Its make-up is
shown in Fig. 2, as are the gauge holes. In making a measurement
the legs of the instrument are inserted in the gauge holes, a read-
ing taken, the instrument taken out and again inserted and read,
and this proceeding repeated until a number of readings without
serious discrepancies are found. The operation of making a

FIG. 3. — TAKING AN OBSERVATION.

measurement is shown in Figs. 3 and 4. The instrument is of
a simple character, but its use requires unusual care and skill on
the part of the manipulator. The method of using the instrument
as well as the necessary general conditions attending such tests
are comprehensively discussed in the paper presented at this
convention by W. A. Slater on Tests of Reinforced Concrete Build-
ings under Load.*

Acknowledgment. — ^These tests were undertaken through the

* See page 168.— Ed.

Digitized by

Google

Beport of Committee on Reinforced Concrete.

eflforts of the Committee on Reinforced Concrete and BuUding
Laws of the Association, in co-operation with the Engineering
Experiment Station of the University of Illinois. The money
to defray the expense of the test was arranged for by the Presi-
dent and Treasurer of the Association. The contractors for the
two buildings co-operated in the tests. The technical part of
making the tests was done by members of the staff of the Engineer-
ing Experiment Station of the University of Illinois.

PIG. 4. — TAIONG AN OBSBBVATION.

Comments, — ^A few words on the basis and limitations of
such tests may not be out of place. The measurements and ob-
servations are subject to some uncertainty; they are not exact
or precise — some erratic readings must be expected. The measur-
ing instrument is used under unfavorable conditions. The gauge
holes are deep in the concrete and the measurements may be
interfered with by dust or other obstructing matter. Great care
and much skill is necessary in making observations. Each test
of this kind made has shown advances in accuracy and certainty^

Digitized by

Google

66 Report of Committee on Reinforced Concrete.

and further experience ought to show further progress. It must
be understood that the structure itself is not entirely homogeneous
and that all parts of it do not act alike. Further, the structure
itself is tied together so closely that stress in one portion may be
modified or assisted by another portion which may not be thought
to afifect it, and this in an unknown amount. The modulus of
elasticity of the concrete in the structure may not be known.
The load-deformation lines may be irregular and imperfect. This

3S £2 05 OQ SS EH ^ fin ir
r^l2ie!I62I!SS!10SB!!Ti3

^ 12 EB SE SI fSi SS BH !S! SQ EH! IS It
■*2?i £2 URUBtSi EESfSSiS! EH !!!!& ^
fTT 177? TTTj mi oil iin im fiii fin iiii if ji

^n pti« Iff if fffp pfF IB" fvn im nn M I ^

FIG. 5. — THE WENALDEN BUILDING.

all means that care must be taken in the interpretation of results.
Important information will be brought out by such tests, as these
tests show, and tests of special features of construction and an
accumulation of data on the action of the structure as a whole
will be worth many times the cost of the work.

Wenalden Building Test.

Building, — The Wenalden Building, Fig. 5, is a ten-story
reinforced concrete structure at 18th and Lumber Streets, Chicago.

Digitized by

Google

Report of Committee on Reinforced Concrete. 67

It was built by the Ferro-Concrete Construction Company,
Cincinnati, Ohio, in accordance with the plans and specifications
of Howard Chapman, architect. It is now occupied by Carson,
Pirie, Scott and Company, dry goods merchants, as a warehouse.

Elevation of Intermediate Beam.

Elevation of Column Beam.

ri^iri^^:i^4;!a

View of Girder.

Fia. 6. — GENERAL POSITION OP REINFORCEMENT.

The building is of the beam and girder type. The floor panels
are 15 ft. by 20 ft. The girders are placed between columns in
the short direction. Floor beams extend the long way of the panel,
there being two intermediate beams built into and supported by
the girders and a column beam built into and supported by the

Digitized by

Google

84 Report of Committee on. Reinforced Concrete.

if^.,4,ij^fer-^',^^-,4^^

II*

'I I
r

^.1

8L-— L-.^ ^-— IJ

□

^ij u u — ' — u u u ir^

FIG. 17.— PLAN SHOWING LOCATION OF GAUGE UNES ON UNDER SIDE OF FLOOR.

Digitized by

Google

Report op Committee on Reinforced Concrete. 85

i 11

i^ 1^ ' *A a44a ^^a: ■» j*»^

^ a

S"?!

jAV

— u u —

FIO 18. — ^PLAN 8H0WIN0

— u u u u \r^

LOCATION OF GAUGE UNES ON X7PPBB BIDE OF FLOOB.

Digitized by

Google

86 Report of Committee on Reinforced Concrete.

Preparation for the Test — ^A week was used in preparing for
the test. Platforms supported by scafifolding for the use of
observers were built on the second floor. Independent of this
was a framework, which was supported by the second floor, for
use in making measurement of deflections. The boxes for hold-
ing the sand were constructed, this being facilitated by a power
saw located on the second floor. Considerable time was con-
sumed in drilling holes in the concrete to bare the reinforcement.

p^J^^^^J^^u,

I L

-!fr-J -

FIG. 19. — LOCATION OF SAND BOXES AND FLOOR CRACKS.

In some cases this was found to be at a considerable depth from
the surface. In all nearly two hundred holes were cut in the con-
crete. Holes were drilled in the reinforcing bars, as heretofore
described, for use as gauge points. The gauge length was made
8 in. The position of the gauge lines for the reinforcing bare
is shown on Figs. 17 and 18 by the even numbers. For use in the
measurement of deformations of the concrete, holes about \ in.
in diameter and 1 in. deep were drilled in the concrete and steel

Digitized by

Google

Report of Committee on Reinforced Concrete. 87

FIG. 20.— VIEW OP SAND BOXES.

ip--»'»f

FIG. 21. — ^VIEW OF TEST LOAD IN TURNER-CARTER BUILDING.

Digitized by

Google

88 Rbpokt op Committee on Reinfobced Concrete.

^00
None ^

O ^ Ok ^ ^ ^ ^ _. . _ ^ _,

%%%%%_%%%%%

no. 22. — LOAD DEFORMATION DIAGRAMS FOR UNDER BIDE OF BEABC8 AT END.

Digitized by

Google

Report of Couuhtee on Reinforced Concrete. 89

Ahne

Do/or-ma'f'lon p^rUnti' o,

^ => § 5 ^ S S
i ^ S § § §

no. 23.— LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF BEAMS AT END.

Digitized by

Google

90 Report op Committee on Reinforced Concrete.

plugs were inserted and set in plaster of Paris. Gauge holes
for receiving the points of the extensometers were drilled in these
plugs with a No. 54 drill. The position of the gauge lines is shown
in Figs. 17 and 18 by the odd numbers. The gauge length was
Sin.

The deflections were measured between a steel ball set in
the under surface of the beam and a ball attached to the frame-
work previously described. The measurements were made by
means of the micrometer shown in Fig. 1.

Method of Loading. — The test area was on the third floor.
The loading material was damp sand which was placed in bot-
tomless boxes. These boxes were of various sizes and were placed
in such a way as to give a well distributed load. The general
size of the box was 4 ft. 6 in. wide, 8 ft. long and 4 ft. 6 in. deep.
Fig. 19 shows the position of the boxes and the test area. Fig.
20 is a view with the sand boxes ready for loading. The boxes
were made small enough to permit a good distribution of load
even though part of the weight of the sand might be carried by
arching and friction down the sides. The test area covered three
full panels and parts of four others, in all equivalent to five panels.
A loading space was chosen which it was thought would give the
fullest stresses over the girders and beams on which the principal
measurements were made. In removing the load the outer panels
were unloaded first in an attempt to determine the relation be-
tween single panel loading and group loading. The load applied
was the equivalent of 300 lb. per sq. ft., double the design live
load.

Before beginning the test, a calibration of the heaviness of
the sand was made by weighing the sand which had been shoveled
into a box of 16 cu. ft. capacity placed on the scales. It was
found that there was a difference of about 10 per cent, in the weight
of sand which had been thrown in loosely and sand which was
packed somewhat. During unloading, the entire contents of
three of the sand boxes (about 500 cu. ft.) were weighed. This
gave an average of 88.6 lb. per cu. ft., agreeing closely with the
weights of the unpacked sand previously weighed, and this value
was used in the calculation of loads.

On a part of the area where the boxes were not carried to
a suflicient height and where the space was not covered adequately
by them, cement in sacks was used as loading material.

Digitized by

Google

Report of Committee on Reinforced Concrete.

The supply of sand for the loading had previously been de-
livered on the same floor, the piles being kept at least one panel
away from the location of the test area, and this was distributed
over sufficient floor space that the stresses in the beams of the
test area could not be affected. In applying the load the sand
was wheeled in barrows and dumped into the boxes. As the
sand was placed, the sides of the boxes were rapped to break the
adhesion of the sand. Some leveling of the sand in the boxes

Abne

None ^

^aoo

^200

125

Qi 15 JVi

£i}L

2J3

FIG. 24. — ^LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF BEAMS AT END.

was done, but there was little compacting by tramping or other-
wise.

Making the Test, — A very important element of a test of
this kind is the initial observation for fixing the zero point of the
test readings. Three sets of observations for a number of gauge
lines were made before the beginning of the test, on the after-
noon of September 10 and the forenoon of September 11.
Where discrepancies were found new observations were made.
Even with this number of observations there are uncertainties

Digitized by

Google

92 Report of Committee on Reinforced Concrete.

Mono

zSOO

^00

£
\AII ^200

\aii \/oo

^^00

^ )EFHt \dO0

\Af/ ^2O0

v%w'^

£ ^300

\£diH^300

\KrHi ^00

^All ^300

%An ^200
5

3*4

31 Z

3ve

390

3U f'C

^ S §

^ ^ ^

JC€

§ ^ « > - « « ^

3f€

t-.

&« b

X
B

FIG. 25. — ^LOAD DEFORMATION DIAGRAMS FOR UPPER SIDE OF BEAMS AT END.

Digitized by

Google

Repokt op Committee on Reinfokced Concrete. 93

in some initial readings. Experience confirms the view that
before any load is placed the initial readings which have been
taken should be worked up and observations repeated untiT all
discrepancies and uncertainties have been removed.

Readings were taken immediately after the completion of each

if ^00

^>^// %200
£ ^^OO

£ ^300
\£6iH^^300

\defhi ^O

'AH

^AH ^2tt?

nQ.26. — ^LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF BEAMS AT MIDDLE.

increment of load and again immediately before the beginning
of placing another increment of load. This usually corresponded
with evening readings and morning readings. A series of readings
was also taken with the full test load on. These extended over a

Digitized by

Google

94 Report of Committee on Reinforced Concrete.

^A// <300

M>n9^

£ ^300

^£0LH^300

^>f// ^300
^An ^20O

JO J

X 5 « ;^

S/S

307

§ ^

3?S

305\

30.'

^ 15 JVi > S>
^ d ^ d ^

§§

FIO. 27. — LOAD DEFORMATION DIAGRAMS FOR UPPER SIDE OF BEAMS ATBODDLI.

£ ^00
^£AH^O

%An '^200

§ S <S

a S g 5 g S §

o C? C5 ^ ^ c[ c>

FIG. 28. — LOAD DEFORMATION DIAGRAMS FOR UNDER SIDE OF GIRDERS AT END.

Digitized by

Google

Repobt of Committee on Reinforced Concrete.

Abne^ 0,

FIG. 29. — ^LOAD DEFORMATION DIAGRAMS FOR UPPER SIDE AND UNDER SIDE OF

GIRDERS AT MIDDLE.

aoo

^A//

A// voo
Uorm

£ ^300
\DEFm

%A!I [^20?

^A/l ^00

Nfme^ Or

§ § ^

/£»/'

/ZO^

/Z7S

;s^ <\| K^ ^ >^ CVl
§ S ^ § ^ ^

/a 03

/k*/S

I S *

FIG.30. — LOAD DEFORMATION DIAGRAMS FOR CONCRETE ON UNDER SIDE OF SLAB.

Digitized by

Google

96 Report of Committee on Reinforced Concrete.

TaBLB IV. — SCHBDULE OF LOADING OPERATIONS IN TuRNBR-CaRTBB

Building Test
Loading Schedule,

Day.

Sunday. . . .
Monday...
TuMday...
Wednesday

ThUTBday..,

Date.

9-10-11

0-11-11

9-12-11

9-13-11

9-14-11

Obeervationa.

Load,
lb. per
sq.ft.

100

200

300

Hours.

lb. per
sq.ft.

12 m

2 p. M.

7.20

12 m.

e.30

8.16 A. M.

6.20

8.20 A. M.

100

200

300

8.00

8.30 A. M.

ling.

ObservaUons.

Load,

Hours.

lb. per
sq. ft.

Hours.

1.30

100

6.10

6.00 p. M.

8.00 p. M.

10.30 A. M.

200

3.10

3.00 p. M.

6.30 p. M.

9.00 A. M.

300

3.60

800

3.30 p. M.

6.80 p. M.

1030

11.30 p.m.

300

3.00

3.30 p. M.

Unloading Schedule,

Friday

Saturday. .
Monday...

Tuesday.

9-15-11

9-16-11

9-18-11

9-19-11

Wednesday.

300

7.30 300 on D. 3.30

to E, F, H, to

I 9.30 A. M. and /. 7.30 p. m.

300 on D. 7.20
if, F. H, to

and J. 9.15 a. m.

300 on £
andff.

9-20-11 300 on ^
I only.

6.15 a.m.

to
9.20 A. M.

8.30 A. M.

Zero.

12.30 p. M.

300 on £ 9.30
and H. to

I 11.45 a.m.

300 on ^ 9.30 a. m.
only. to

* 12.00 m.

1.00

3.40 p. M.

300 on Z),

E, P, H,

and I.

300 on £
andff.

3O0onE
only.

300 on £
only. I

Zero on
all bays.

8.00

8.30 p. M.

6.30

8.00 p. M.

12.16

1.60 p. M.

4.16

8.00 p. M.

4.60

6.60 p. M.

4.00

6.40 p. M.

Digitized by

Google

Report of Committee on Reinforced Concrete. 97

period of 48 hoxirs. A similar method was used in the process of
removing the load.

Table IV shows the loading schedule. The load was applied
in increments of 100 lb. per sq. ft. based upon the whole test area.
The application of the load consumed three days. The full load
was left on 48 hours. The unloading schedule is also shown in
Table IV. In the unloading, the load on panels B and C were
first removed, then the load on panels Z), F, and 7, followed by the

\300

\200

voo

1-4//

Nbne O

FIG. 31

-LOAD DEFORMATION DIAGRAMS FOR CONCRETE ON UPPER SIDE OF SLAB.

removal of the load on panel H, Fig. 21 is a view at a load of
300 lb. per sq. ft. over the test area. The total load was over
500,000 lb.

Personnel of Testing Staff. — ^All instrument readings were
made by W. A. Slater and H. F. Moore, of the stafif of the Engi-
neering Experiment Station of the University of Illinois. Mr.
Slater had immediate charge of the test as a whole. A. N. Talbot
was present during the work of preparing for the test. Three
others assisted in the work of recording and reducing data.

Digitized by

Google

98 Repokt op Committee on Reinforced Concrete.

Table V.— Strbss Indications in Turner-Cartbh Building Test.

Member.

Gauge Line.

220
244
304
318
310

202
206
230
234
236
238
240
222
224
214

Reinforcement.

Gauge Line.

269
311

265
267
281
293
301
305
313
315

Concrete.

End of girder

Middle of girder

Rnd of beam

'siooo

9,000
8,000
8,000
4,000

"7]666

11,000
9,000
8,000
8,000

11,000

5,000

—3,000

900
Little

' 1,166

1,100

1,000

800

Middle of beam

350

200

300

Rent up bar in girder

Bent UD bar in beani

Table VI. — Maximum Stresses and Moment Coefficients
in Turner-Carter Building Test.

Member.

Girder, End —
" End....
" Middle .
" Middle.

Intermediate Beam, End —
End....
Middle.
Middle.

Column Beam, End. . . .
End....
Middle.
Middle.

Reinforcement.

Btreea.

31,000

ik'ySdo

8,000

Sl,500

8,000

18,600

11,000

lOfiOO

17,000
10,000

Coefficient-

1/lB

1/12
0.05

1/12
0.03
1/12
0.05

1/12

1/12
0.05

Concrete.

Stress. Coefficient.

IfiOO
900

too

Little

1^00

1,100

S80

350

1,200
950
S50
225

1/12
0.06
1/12

1/12
0.07
1/12
0.077

1/12
0.064
1/12
0.054

Digitized by

Google

Report of Committee on Reinforced Concrete.

Deformations and Stresses. — The results of observations on
various gauge lines for the beams and girders are plotted in Figs.
22 to 29. Fig. 30 gives the deformations in the concrete on the
under side of the floor slab and Fig. 31 those on the upper side
Fig. 32 records measurements made on the bent-up bars and
stirrups.

As already stated, the location of the gauge lines is shown on
Figs. 17 and 18, the odd numbers referring to measurement on the

S^300

thne

Ahne

200

too

E ^300
\CdLH^O

\defhi \zoo

^AN ^00

^All ^2tt7

^AH %00

None^ 0,

^(£.

/i^^^

^'S

^/4-

Zi*C

^^a

2/'^l

De "fonma -f-ion per Un // or L en^-f-h.

no. 32. — ^LOAO DEFORMATION DIAQRAMS FOR BENT-UP BARS AND BTIRRUPe.

concrete, the even numbers to measurement on the reinforcement.
The numbers in the two hundreds are gauge lines on the under
side or second story side, and the numbers in the three hundreds
are on the upper side or third story side.

Stresses and bending moment coefficients are tabulated in
Tables V and VI.

The suggestions given for caution and care in interpreting
measurements should be applied to this test.

Digitized by

Google

100 Report of Committee on Reinforced Concrete.

Beams. — For the tensile stresses in the reinforcement at the
middle of the intermediate beams at the full load of 300 lb. per
sq. ft., the highest stress observed ^as 11,000 lb. per sq. in. and
the average stress recorded may be said to be 8,500 lb. per sq. in.
At the ends of the intermediate beams, the highest stress observed
in the reinforcement was 8,000 lb. per sq. in., and the general
value may be said to be 7,500 lb. per sq. in. Using the assump-
tions for resisting moment ordinarily taken in design calculations,
these stresses may be considered to correspond to a bending

^aioffce from C0k/m/7-j

•/nch^iS

^ § 35 5 > 5k

no. 33. — ^DIAGRAM SHOWING DISTRIBUTION OF COMPRESSIVE DEFORMATION IN
BOTTOM OF COLUMN BEAM.

moment coefficient of .05 Wl for the maximum stress at the middle
of the beam and .03 Wl for the maximum stress at the end of the
beam, if the tensile strength of the concrete be not considered.
Assuming a modulus of elasticity for the concrete of 2,500,000
lb. per sq. in., the concrete on the compression side of the beams
at the middle showed a compressive stress of 350 lb. per sq. in.
and at the end of the beam 1,100 lb. per sq. in. It is apparent
that the total compressive stress in the concrete is greater than
the total tensile stress in the reinforcement of the beams. A
possible explanation is that end thrust exists, involving so-called

Digitized by

Google

Report op Committee on Reinforced Concrete. 101

arch action in the beams and floor structure, and that the tensile
stress is relieved by the presence of this thrust. The tensile
strength of the concrete must have a large effect on the resisting
moment. The coefficient of Wl in the bending moment, necessary
to give a compressive stress equal to the maximum measiu^ in
the concrete, on the assumptions made, is .077 for the middle of
the beam and .07 for the end of the b^am. These coeflScienta
are lower than the value 1/12 usually assumed in design of such
beams.

Girders, — For the tensile stresses at the middle of the girders
the observations showed about 8,000 lb. per sq. in. in the reinforce-

' O O ' - '

o o o

whncofrvm y/rder' inches ^

FIG 34. — DIAGRAM SHOWING DISTRIBUTIGN OF COMPRESSIVE DEFORMATION IN
INTERMEDIATE BEAM.

ment at the middle. This corresponds to a bending moment
coefficient of .05, again neglecting the tensile strength of the con-
crete. The reinforcement at the end of the girder was inaccessible.
Assuming a modulus of elasticity of 2,500,000 lb. per sq. in.,
the concrete on the compressive side of the beam at the support
showed a compressive stress of 900 lb. per sq. in. The reading
at the middle of the beam showed very little compression. Assum-
ing that the loads on the girder are concentrated at the points
where the intermediate beams are connected, and making the same
assumption of distribution of stress as before, the coefficient of
bending moment was .06. It seems probable that the compres-
sion at the middle of the span must be distributed over a consid-

Digitized by

Google

102 Report op Committee on Reinforced Concrete.

erable width of floor, or larger readings of compression would have
been obtained.

Decrease in Compression with Distance from Support. — ^In four
beams measurements of compressive deformations were taken at
a series of gauge lines from the support to a location near the point
of inflection. The position of these points is shown in Pig. 17.
The gauge lines No. 223, 225, 227, 229, 231, and 233 are on one
side of column No. 6, and 281, 283, 285, 287, 289, and 291 are on
the other side of colxmin No. 6. It may be expected that there
will be full restraint for the end of the beams. Gauge lines 243,
241, 239, 237, and 235 are on one side of a girder and 293, 295,
297, 299, and 1201 are on the other side. The unit-deformations
for these gauge lines at loads of 200 lb. per sq. ft. and 300 lb. per
sq. ft. are plotted in Figs. 33 and 34.

The measurements recorded for the column beams show con-
siderably more compressive stress than do those for the intermediate
beams, perhaps one-third more. This difference in stress may be
due partly to the deflection of the girder, and to the deflection of
the intermediate beam between its support and a point opposite
the end of the column beam, which would permit a larger part
of the load to be carried by the column beam. It may be due
somewhat to the fact that reinforcing bars are bent down from a
point at the end of the column beam, while in the intermediate
beams the bars run horizontally for a foot from the face of the
girder.

The direction of the lines in Fig. 33 and Fig. 34 indicates a
zero stress at about 45 in. from the face of column in the column
beams and at about 50 in. from the face of the girder in the inter-
mediate beams. In both cases the results locate the point of
inflection at about 0.22 of the clear span.

T'Beam Action. — The distribution of compressive stresses
in the T-beam formed by a beam and the floor slab (which involves
the distances away from the beam for which compressive stresses
are developed) has been a fruitful source of discussion. Measure-
ments parallel to the axis of the beam were taken on the upper
surface of the floor slab immediately above beams and at intervals
between them. These gauge lines are No. 315, 317, 319, 321, 323,
325, and 327 (see Fig. 17). The deformations are shown in Figs.
27 and 31. The amount of these deformations at points across the

Digitized by

Google

Report of Committee on Reinforced Concrete. 103

slab for loads of 200 lb. and 300 lb. per sq. ft. is shown in Fig. 35.
It is apparent that a somewhat higher stress existed in one beam
than in the other. Taking this into consideration, the compres-
sive stress varies quite uniformly from one beam to the other, and
the full width of the floor slab may be said to be effective in taking
compression. The overhang (counting to the midpoint between
beams) is 6} times the thickness of slab. It will be noticed that
the conclusions are the same as given for the Wenalden building
test.

Readings were also taken on the under side of the floor slabs
parallel to the beams at three places (No. 1205, 1211, and 1213),
but the conditions attending the location of these points do not
permit conclusions to be drawn.

Floor Slab. — Measurements were taken on the floor slab in

.0002

■_

^^—

"^o""

.ooot
o

—

— ,

--—

—

""">«

- —

-^^i2/^

:; ^ « B! s

no. 35. — DIAQBAM SHOWING DISTRIBUTION OF COMPRESSIVE DEFORMATION
ACROSS FLANGE OF TEE BEAMS.

the direction of its span at three places on the under side and at
one place on the upper side immediately above one of the lower
measurements. These gauge lines were Na 277 on the under side
of the slab close to a girder (Fig. 17), No. 279 on the under side of
the slab 5 ft. from the edge of the girder, No. 309 (Fig. 18) on the
upper surface immediately above No. 279, and No. 1203 (Fig. 17)
on the under side half way between girders. The measurements
are plotted in Figs. 30 and 31. As might be expected from being
close to the girder and near the level of its neutral axis, No. 277
showed little deformation. The pair of gauge lines (No. 279 and
309) shows less deformation than would be calculated by the ordi-
nary beam formula, but perhaps not less than would be the case
if the tensile strength of the concrete is considered to be quite
effective. The reading of No. 1203 was even smaller than 279.

Digitized by

Google

104 Report of Committee on Reinforced Concrete.

All the stresses found in the floor slab were low. The deformations
parallel to the beams were discussed under T-beams.

Bond Stresses, — ^At the ends of the beams the reinforcing bars
lapped over the center line of the girder a distance of 15 in. An
effort was made to determine whether there was a movement of
one of these bars with reference to the adjoining concrete and with
reference to the adjoining bar; also whether the deformation in
the stub end of the reinforcing bar was the same as in the adjoin-
ing bar. Fig. 36 shows the location of the reinforcing bars with
reference to each other, and the position of the gauge lines. No.
312-14 in comparison with No. 312 and 314 will indicate any
relative movement of one bar with respect to the other, and No.
312c and 314c in comparison with No. 312 and 314, respectively
will indicate any movement of the bars with respect to the con-
crete.

It appears possible that the initial reading of No. 314 is

Jrife/ plug m concrt*f^

no. 36. — ^ARRANQEMENT OF GAUGE POINTS TO TEST FOB MOVEMENT OF BAR
RELATFVE TO CONCRETE.

slightly in error, and the remarks already made about quantitative
interpretation of results and the chances for variations in stresses
in adjacent bars or in adjoining concrete should be borne in mind
in studying the results. It seems evident that No. 314 (on the
lapped bar) records considerable less stress than No. 312. The
measurements indicate a possibility that the right-hand point
of gauge line No. 314 has moved to the right relatively to the right-
hand point of No. 312, though this amount may not be more than
the amount of initial slip necessary to develop the requisite bond
stress. The measurements taken have no bearing on whether
the left-hand point of No. 314 has moved. The measurements
also indicate that there was no motion of the left-hand point on
the reinforcing bar (No. 312 gauge line) relatively to the concrete
at its side, though it must be borne in mind that the point taken
was so close to the bar that only slip and not distortion of concrete
could be measured.

Digitized by

Google

Report of Committee on Reinforced Concrete. 105

Web Deformations, — No diagonal tension cracks were visible
on any of the beams or girders.

In girder 4 measurements were taken on the diagonal portion
of a reinforcing bar, one of the bars which is provided to take
negative bending moment. This is shown in Fig. 17, Section K-K.
The gauge lines are No. 222, 224, and 226 The position of the
gauge lines is also shown in Fig. 16. The measurements are plotted
in Fig. 32. It was impracticable to measure the deformation at
a point closer to the support. The measurements show about
the same stress at No. 222 and 224, perhaps 5,000 lb. per sq. in.
The stress at No. 226 is materially less. It is not improbable
that there was tension in this rod throughout its length. As there
was considerable compression measured in the gauge lines on the
bottom of the girder below No. 222, it seems probable that a
crack was formed in the top of the floor slab somewhere above No.
222, but as this space was filled in with bags of cement no observa-
tion was made during the test, and inspection of this space after
the load was removed seems to have been overlooked. At the
other end of the girder, near column 6, a fine test crack was found
on the upper surface of the floor 2 in. from the face of the column
extending across the width of the girder and beyond. This
extended through the floor. A similar crack was observed on
girder 3 near column 15.

Gauge line No. 228 is on a stirrup (see Fig. 16). This stirrup
is in an inclined position. It is not known what bar it is intended
to be connected with, nor whether there. is connection with a
tension bar. The gauge line is in a region of the beam where hori-
zontal compressive stresses may be expected. The measurement
in the stirrup at the first increment of load shows tension (see Fig.
32) and subsequent increments give compression. It should be
noted that readings could not be taken on the upper end of the
stirrup. If the upper ends are merely bent out into the floor slab
it is hard to see that the stirrup may be expected to be useful in
transmitting web stresses.

In beam 9 (see Fig. 17, Section L-L, gauge line No. 218)
measurement was taken on the diagonal portion of a reinforcing
bar which is carried through the girder at its top and a few inches
beyond. See also Fig. 16. This shows a tension of 3,000 to 5,000
lb. per sq. in. (See Fig. 32.) This bar was inaccessible from the

Digitized by

Google

106 Report of Committeb on Reinforced Concrete

Non9

\DEnff \$00
'^A/t <5O0

\a// \lOO

^ 8 § S % ^

Def/ecr/on^ /n inches.

FIG.— 37. LOAD DEFLECTION DIAGRAMS.

Digitized by

Google

Report op Committee on Reinforced Concrete. 107

top of the floor, but the gauge lines on the companion bar (No.
324 and 318) show about 5,000 and 9,000 lb. per sq. in. Measure-
ments in the diagonal portion of a single-bend bar (gauge lines
No. 216 and 214, Fig. 17) which extends only to the center of the
supporting girder indicate a small compression in the bar (see
Fig. 32). A stirrup, which like the one in the girder was close to
the end of the beam and was inclined so that its lower end was
nearer the support than its upper, showed shortening of the stir-
rup (see gauge line No. 212, Figs. 16, 17, and 32). In both cases,

FIQ, 38. — CABINET PROJECTION SHOWING BEAMS AND GIRDERS AND POSITION OF

TEST CRACKS.

the arrangement was such that the stirrup could hardly be effec-
tive.

The amount of the vertical shear in the beams and girders
was such that diagonal tension cracks might be expected except
for the small tensile stresses in the top of the girder and the end
constraint which seems to have been developed in both beams and
girders.

Deflections. — ^The deflections of the beams (including that due
to deflection of girder) and the deflections of girders are given in
Fig. 37. The location of the deflection points is shown in Fig. 19.

Digitized by

Google

108 Report of Committee on Reinforced Concrete.

The effect of time upon the deflection is shown by the increase
in deflection under constant load. The change when portions
of the load had been removed may be due in part to the time
element and in part to the effect of location of the load on the
panels. The deflections seem relatively small, especially when
compared with deflections obtained in laboratory tests of beams
carrying the same loads. The effect of the time element is indi-
cated on these diagrams. The conditions were such that the sup-
ports were subject to possible displacement by workmen.

Columns. — Readings were taken on the four faces of Column
No. 6 just below the girders, but the results are not consistent
enough to warrant attempting drawing conclusions.

Test Cracks. — Fine tension cracks were observed in the lower
part of the beams and girders. The location of the observed
cracks is shown on Fig. 38. The appearance of these fine cracks
is similar to those observed in laboratory tests. They would
not be noticed without specially careful examination.

The floor cracks already mentioned indicate the development
of the tensile stresses in the beams and girders at the support.

The limitation of space and time have prevented the pre-
sentation of other matters which were observed in the tests.
For example, the observations on deformations during the 48
hours time with the full loading showed in general a slight increase
in the deformations in the reinforcement and in the concrete.
It is hoped to take up some of these matters at another time.
It was not possible to give full attention to every feature upon
which information was sought, and in some cases isolated points
were used with a view of determining tendencies, and in these
naturally there is less certainty in the indications.

Part II. Test of a New Type of Flat Slab Floor
Construction.

Building. — The Powers Building, Fig. 39, is a three story
and basement warehouse located at Minneapolis, Minn. The
exterior walls of the building are bearing walls and there is one
row of columns through the middle of the building. The floors
are flat slabs, reinforced in two directions with high elastic limit
deformed bars. Fig. 40 shows a basement and first floor plan

Digitized by

Google

Report op Committee on Reinforced Concrete. 109

of the building, Fig. 41 the reinforcement in the floor tested and
Fig. 42 the reinforcement in place.

General Outline of Test — ^The first floor of the building was
selected for the test by the Building Inspector of the City of
Minneapolis, as this represented an acceptance test for the new

FIG. 39. — POWERS BUILDING, MINNEAPOLIS, MINN.

design. At the time of the test the floor was about 3 months old.
Four panels were loaded to 200 lb. per sq. ft. (the design load)
and then two panels were loaded to twice this amount. Cement
in bags was used for the load, piled in piers to prevent arching.
The load was arranged as shown in Fig. 44 until the design load
was reached, while for the maximum load Panels 2 and 3 only

Digitized by

Google

no Report of Cobimittee on Reinforced Concrete

.J^

«Si

open^

i Bl

ytasement Columns • Zo'square %

first ^ory Columns- id'square.

io\o'

7%^ Rough 3lab except
Rectangles over Columns
which are iwo inches
1 thicker

_ jji^Cemenf Finish

-4 Tile Partition in
first Story only.

— f —

l_.j:— J

•H

^..

/9'-6'

no. 40. — PI.AN or powers building, baskmxnt and rbbt floob.

/Google

Digitized by ^

Report of Committee on Reinforced Concrete. Ill

dors extend ^ same dirsction as lettering indicating them ■ straight A bent
bars alernate. - StmigHt bars extend 6' past margin of panel ana bent bars
extend to quarier-jpoinr ofadjdcenf panel

There one onfy ho layers of bars m the top of the siaband two ^
vss\ lowers in the bottom. The bent bars extending across the building fbrm%
^ the bottom tiyer in the middle oftfjepanel and the top layer over the

the pat
Mm

i^&QBlMBi

TIQ. 41. — DIAGRAM 8H0WINQ ARRANGEMENT OF REINFORCEMENT.

Digitized by

Google

112 Report of Committee on Reinforced Concrete.

were loaded. The wide aisles shown were necessary for the
accommodation of the mstruments and observers. All tabulated
loads per square foot and the loads used in plotting are in every
case the total load on the panel divided by the area of the panel;
the intensity of the load under the piers, of course, is greater.
Deflections of the various panels were measured under different
stages of loading and also the deformation of the reinforcement
and concrete due to these loads.

FIG. 42. — REINFORCEMENT IN PLACE, POWERS BUILDING.

Instruments. — The deflections were measured with a deflec-
tometer, see Fig. 1.

Deformations in the reinforcement and concrete were meas-
ured with an 8-in. Berry strain gauge. Fig. 43, which reads direct
to 1/2000 of an inch, or by estimation to one-fourth of this
amount. As slight variation is possible, 5 readings were taken
at each point and the average of these assumed to be correct.
Readings were taken at intervals throughout the test on stand-
ard bars and on standard points placed in imstressed portions of
the concrete; the temperature corrections so observed have been

Digitized by

Google

Report of Committee on Reinforced Concrete. 113

made to the readings. There are many diflSculties mvolved in
measuring deformations under conditions such as exist in a test
of this kind. In view of these difficulties extreme accuracy can-

Frame

'Ame6 Gauge

1 1

■Plunger I

Lever

Hxed Leg

^35 Hole

^55 Hole

no. 43. — BEBBT STRAIN QAUOE.

not be hoped for, but serviceable results can be obtained by a
trained observer. Observations were made by W. A. Slater,
University of Illinois, and F. J. Trelease of the Research Depart-
ment, Corrugated Bar Company.

Digitized by

Google

114 Report of Committee on Reinforced Concrete.

FIG. 44. — ARRANGEMENT OF LOADING.

Digitized by

Google

Report of Committee on Reinforced Concrete. 115

Application of Load, — ^The test continued for 9 days, July 27
to August 4, inclusive, 1911. The loading was in 10 stages as

Table VII. — Schedule of Loading Operations.

Stage.

Load ic

Place

Loadio

Poiinds

per Panel and per Square
Foot.

From

Panel 1.

Panel 2.

Panel 3.

Panel 4.

1
2
3

4
5
6

1.45 P.M., 7-27
11.10 A.M., 7-27
4.45 P.M., 7-28
10.45 A.M.. 7-29
3.40 P.M.. 7-29
7.00 P.M.. 7-29
3.15 P.M.. 8-1
8.20 a.m., 8-2
4.30 P.M.. 8-2
10.35 a.m.. 8-3

8.40 A.M., 7-28
2.00 P.M.. 7-28
8.45 A.M., 7-29
1.25 P.M.. 7-29
5.40 P.M.. 7-29
11.00 a.m.. 7-31
4.25 P.M.. 8-1
9.15 a.m.. 8-2
7.15 A.M.. 8-3
3.45 P.M.. 8-3

13,300
33.440
53,200
34.200
15,200

52
132
210
135

13.300 51
33,440 128
53.200 204
72,200 276
91.200 1 350
106,400i 408

L^.iiLtO
3^S.].|.0
f.:j,JiiO

r-'.-iiO

ItKLiOO
lOf-MK)

Sa.Ui'O

51
129
206
280
354
414
198

13,300
33.440
53,200
34,200
15.200

51
128
204
131

96,900 1 370
106.780' 408

53,010
21.660

202
83

Note.— Started placing Stage 1 at 10 A. M.. 7-27-1911. All load removed at 4.20 P. M..
8-3-1911.

will be noted from Table VII, which shows the arrangement of the
load at various stages and the length of time each load remained on

FIG. 46. — DESIGN LOAD IN PLACE ON FOUR PANELS.

the floor. Readings were first taken on all points with the floor
unloaded and then a load approximately equivalent to 50 lb.

Digitized by

Google

116 Report of Committee on Reinforced Concrete.

■X

•7

• 6

•5

•4

• J

• 2

'^ r

! L

I I

•/J

lj2is^

•iz

•10

— -rf/7

i/4

I I

•<S

;-— ^

•27

• 26

• 25

•24 •3?

• 22 •SO
•21 ^29
•20 ^28

• 19

:l . J

FIG. 46. — ^LOCATZON OF DEFLECTION POINTS.

Digitized by

Google

Report op CoMBniTEE on Reinforced Concrete. 117

per sq. ft. was applied over 4 panels, Fig. 44. Another set of
readings was taken and the load increased to 125 lb. per sq. ft.
This method of alternate loading and reading obtained through-
out the test. When the desigp load of 200 lb. was reached, how-
ever, the load on Panels 1 and 4 was moved by stages to Panels
2 and 3, so that finally Panels 1 and 4 were completely unloaded
and Panels 2 and 3 were loaded to about 400 lb. per sq. ft. The

Table VIII. — Deflections in

Inches.

Set

6 7 1 8 ' 9

10 1 11

Load

3 4 5 1 6

6 1 6

-.005
.005
.014
.004
.000
-.002
-.001
-.016
.004
.038
.000
.001
-.001
.008
.016
.017
.012
.010
.013
.022
.021
.020
.028
.019
.016
.008
-.001
.014
.013
.015
.017
.018

.000
.015
.026
.034
.005
.010
.007
-.003
.012
.043
.000
.002
.012
.022
.019
.028
.023
.019
.028
.035
.033
.037
.038
.041
.027
.021
.006
.023
.022
.018
.029
.031

.004
.032
.057
.055
.003
.012
.008
.000
.017
.039
.006
.005
.015
.051
.054
.067
.060
.049
.068
.085
.081
.079
.090
.106
.069

!oio

.061
.169
.065
.079
.078

.Oil
.052
.079
.066
.008
.010
.003
.009
.019
.046
.014
.009
.013
.111
.126
.142
.133
.118
.137
.205
.213
.224
.236
.249
.154
.321
.021
.158
.176
.179
.196
.196

' ; '
; , 1

.016' .026' .043 .065
.060 .091 .143, .197
.090 .139, .217 .294
.073, .102 .144 .195
.013, .0171 .035' .050
.008 .010 .015 .020
.003 .005, .0061 .009
.0121 .007 .002:-. 002
.021: .028 .037, .046
.056, .067' .076 .101
.014, .020' .033' .035
.009, .011] .009' .008
.013' .013, .0151 .015
.126 .134, .1631 .152
.146' .163' .182. .209
.155 .175' .220; .242
.149; .163' .190' .198
.138 .141 .144' .137
.1681 .1631 .1451 .136
.238 .251, .2731 .283
.255, .278 .320| .348
.264' .294' .346, .374
.278 .303' .326 .363
.2861 .289 .318' .316
.156 .179 .1661 .148

.341! .326, .3061

.0291 .034 .027, .023
.1841 .196| .2061 .229
.211, .2301 .2661 .291
.214 .234 .273 .305
.228' .248 .285 .309
.229, .236 .256 .268

.082
.237
.351
.243
.063
.028
.014
.000
055

.07.

•:«4

.070

.059
.174
.271
.196
.070
.026

.040
.112
.159
.102
.047
.012
.010
-.003
.031
.070
.023
.006
.011
.120
.194
.210
.177
.136
.130
.272
.327
.350
.350
.313
.149

.040
.110
.159
.108
.050
.016

-!6d6
.030
.066
.030
.007

".lis

.187
.207
.169
.130
.131
.260
.308
.333
.312
.304
.151

.042
.114
.161
.113
.048
.016
.014
.003
.026
.059
.026
.007
.015
.095
.156
.171
.141
.110
.120
.226
.265
.284
.286
.269
.130

'!i55

.OOi

-.002

10
11

.112
.049
.009
.022
.135
.216
.240
.205
.147
.132
.307
.378
.413
.401
.349
.148
.137
.027
.243
.319
.341
.342
,287

.120

.104

.056

12
13

.010

.008

.009

.131

19
20
21

.148

.161
.320

.282

22
23

.421

.424

.214

24
26
26

'!i49

.363
.156

"iw

.022
.214
.287
.299
.302
.270

";266
.268
.284
.290
.261

.021
.179
.242
.254
.255
.230

28
29

load was next removed from Panel 3 and finally from Panel 2,
readings being taken at intervals during the unloading. Fig. 45
shows the design load in place on the four panels.

Data. — ^Deflection readings were taken at 32 points. Fig. 46.
The deflections are given in Table VIII and are plotted in Figs.
47 and 48. The load plotted in these figures and in the stress
curves is the load per square foot on the central panels. The
stage of loading can be obtained by referring to Table VIII.

Digitized by

Google

118 Report of Committee on Reinforced Concrete.

6 aJ ai o 37 oB aS o ST a? 53"
Deflection In Inches

FIG. 47. — DEFLECTION CUBVK8.

Digitized by

Google

Report of Committee on Reinforced Concrete. 119

Deformation of the reinforcement was read at 30 points and
of the concrete at 31 points. The location of these points is
shown in Figs. 49 and 50. Table IX gives the embedment to
center of bar; thickness of the slab and finish coat; net thickness

Loaded Portion

^; ^Me■»> q J>vlM>« LWA^a^J;Jft^;.w;^■l!<^aw■i^^^^^^^^^

FIG. 48. — SECTIONS OF FLOOR SHOWING DEFLECTIONS.

of the rough concrete slab, deducting the thickness of the finish
coat from the total thickness of slab, as the finish coat was found
to be loose in many places and did not have much load-carrying
capacity; and also the effective depth of the reinforcement on
which the strain was measured.

Digitized by

Google

120 Rbfobt of Comhtttee on Reinfobcisd CoNCBinv.

o4«

"H r

\^-

I I

III-

ItO

116

?05

.„:r-.j

/a5

All Oauge Length6 - (S"'

%I07

?oer

•-• Indicates Points on Top of 5lab

o-o Indicates Points on bottom of 5lab

no. 40. — LOCATION OF DEFORMATION GAUGS POINTS OR REINFORCEMENT.

Digitized by

Google

Report of Committee on Reinforced Concrete. 121

1 — "

•

—

_L_.

!
1

I —
1

m "^ ''^

1
1

4<»>

I 1

1
1
I

415-^ 405^
J3/5 .,p

305

* 304
1

All Gauge Lengths

1
I
1

^311 307

r-'

•-• Indicates Points on Top of J lab

o-o Indicates Points on dottom of 5lab.

nOt. 60.— LOCATION OF DBFORMATION GAUGE POINTS ON CONCRETE.

Digitized by

Google

122 Report of CoMBniTEE on Reinforced Concrete.

The observed elongations and corresponding unit stresses,
based on an assumed modulus of 30,000,000, in the reinforcement
are given in Table X. Unit reinforcement stresses are plotted
against the load per square foot on Panel 2 in Figs. 51 and 62. A
series of readings taken by Mr. Slater on one of the bars over the
colunm cap shows that a bond stress of 135 lb. per sq. in. under
maximum loading existed at that point.

Table IX. — Dimensions at Deformation Gauge Points.

Embed'-

Thickness of

Net Thickness of
Slab, in in.

Effective Depth
of Bars, in.

Gauge -
Point.

ment of
Bar to

Slab.

Finish

From

center, in.

in.

Coat. in.

Actual.

Plans.

Actual.

7V.

Plans.

101

IVi

9Vs

1 V4

7V«

7«/4

7»i,*

102

OVa

7Vi

8V.

8V4*

103

l"/if

9Vt

7Vi

7Vi«

7 v.*

104

IVm

91.

• «

7Vt

*•

7»»/i«

8V4*

105

IVt

8Vs

••

7 V.

7V.

7Vi»

106

1V4
1 "/!•

107

"difii"

"sv;."'

;•

'■7Vi""

"Viu*"

108

1 Vt

109

2Vi«

"sifV"

•*

"7V."

*•

"6«A.' ■

"7ij\i"

111

l»/w

8»»/w

•'

7 "/!•

•*

7V.

7V.«

112

IVw

7V4

7Vw

8V4*

118

1 »/!•

0»'w

114

"svV"

"7V«"*

" * * * "

"7»/i;"

"svii"

115

lu/is

116

IVs

118

1V4

"s'^'lu"

••

**7»Vii"

••7U/1'."

"SVii"

120

OVw

l>/»

122

"si/V"

*"*"'"'

■'7Vi "

"Vvi"

■'s 1/4* '

201

3Vw

lOV.

IV4

9V.

"ovi'"

7Vi«

8».

202

3Vi

lOVi

l»/4

9V.

9V4

7V4

203

2V8

9V.

n/4

7V.

7V4

6Vi

204

2Vw

11 Vi«

l»/4

9>*/i«

9V4

8V«

205

3»/»

10 v»

1V4

9 v.

9 "A

7V.

8Vi

206

3Vi«

10 "/w

1 Vi

9Vi«

9«/«

7Vs

207

2Vi

11 1/4

10 V4

9V4

8Vi/

8V«

208

2>/i«

8"/m

IVii

7Vb

7V4

6Vt

200

8 1/4

l»/4

7V4

5V4

6V«

210

2 Vie

91/1.

l»/l«

7V4

6Vt

* Assuming 1 \ in. finish coat.

Table XI gives the observed elongations and corresponding
miit concrete stresses. These concrete stresses are based on an
assumed modulus of 2,000,000 and are plotted against loads per
square foot in Figs. 53 and 54.

A summary of the unit stresses is given in Table XII, the
stresses giveii being the maximum probable values.

During the test, cracks were carefully searched for with an
electric light and recorded as found. Their location is shown
in Fig. 55. Most of these cracks were very minute and difficult

Digitized by

Google

Report op Committee on Reinforced Concrete. 123

Tablb X. — Deformations in Inches and Reinforcement Stresses in

Lb. per Sq. In.

Set

1
3

LoMl

Gauge
Point
101

102

103

104

105

0.0
1685

0.3
560

-0.3
-560

-0.5
-040

0.4
750

•0.7
1310

0.8
1600

-0.5
-040

0.3
560

1.2
2250

0.8
1500

0.5
940

0.4
750

0.3
560

-0.3
-560

-0.6
-1120

-0.7
-1310

-0.8
-1600

-0.5
-940

-0.2
-380

0.3
560

-0.1
-190

-0.3
-560

-0.3
-*60

0.1
190

-0.6
-1120

0.2
380

-0.5
-940

-0.5
-040

-0.9
-1690

-0.3
-«60

-0.2
-380

-0.3
^60

1.0
1880

1.1
2060

05
940

-0.1
-190

0.6
1120

-0.9
-1690

-0.4
.750

0.5
040

-0.1
-190

0.6
1120

-0.4
-750

0.3
560

-0.4
-750

0.1
190

-0.5
-940

0.1
190

-0.5
-940

-1.6
-^000

-0.4
-760

0.9
1690

-1.0
-1880

0.3
560

-0.1
-190

-0.2
-380

-0.9
-1690

0.5
940

0.2
380

3.3
6200

-0.1
-190

8.6
6740

-0.5
-940

0.2
380

0.0

•••••

0.7
1310

4.2
7880

0.9
1600

4.7
8810

1.1
2060

4.3
8060

2.7
5060

5.0
9380

1.6
3000

4.3
8060

2.7
5060

5.7
10690

2.0
3750

4.2
7880

3.5
6560

6.0
11250

2.2
4120

4.4

8250

4.2
7880

7.0
13120

2.4
4600

6.6
10500

4.8
9000

7.7
14440

1.9
3560

4.0
7500

3.9
7310

5.7
10690

-1.0
-1880

1.6
3000

3.0
5620

3.4

6380

4.9
9190

-0.3
-560

2.8
5250

2.7
5060

2.0
3750

3.7
6040

-1.2

::::: ;::::

-2250

106

1.5
2810

8.0
15000

0.6
1120

0.8
1500

107

0.0

-0.5

-940

106

0.3
190

0.2
380

-1.8
-3380

0.2
380

5.3
9940

-0.6
-1120

1.2
2250

-0.8
-1500

0.2
380

-0.6
-1120

2.8
5250

1.7
3190

1.5
2810

1.6
3000

7.0
13120

0.6
04G

0.3
560

0.9
1690

-2.1
-3940

-0.8
-1500

1.8
3380

-2.4
-4500

-3.1
^10

2.4

4500

-3.0
-5600

0.3
560

.20
3750

-2.5
-4600

109
110

2.2
9120

0.8
1500

1.2
2250

111

-0.3
^60

5.9
11060

5.6
10500

-1.6

-3000

112
113

•••••

6.3
11810

0.4
750

2.1
3940

0.3
560

0.7
1310

-0.7
-1310

2.9
5440

5.4
10120

1.2
2250

2.7
5060

0.9
1690

1.3
2440

-1.4
-2620

1.9
3560

6.0
11250

0.3
560

1.5
2810

0.1
190

1.5
2810

-1.5
-2810

2.4
4500

6.1
11440

1.4
2620

7.6
14250

4.0
7500

114

2.0
3750

1.2
2250

1 7

3190

115

-0.1
190

1.7
3190

-0 R

116

1.2
2250

0.8
1500

117

-040

118

2.1
3940

2.7
5060

1.3

2440

120

1 0.7

1310

0.2

1 380

0.8 1.2
1500 2250

0.4 2.7
750, 5060

-0.4i -0.1
-750 -190

0.5

940

122
201
202
203

1.5
2810

7.6
14250

-0.1
-190

1.4
2620

2.0
3750

8.1
J5190

0.1
190

3.0
5620

1.8
3380

9.2
17250

-0.8
-1500

3.9
7310

3.1
5810

9.1
17060

-0.5
-940

4.0
7500

3.7
6940

10.1
18040

-0.6
-1120

3.4
6380

3.1
5810

9.1
17060

9.5
17810

1.8
3380

2.8
5250

7.8
14620

2.1
3940

2.4
4500

5.8
10880

2.4
4500

2.5
4690

5.2
9750

-0.2

Digitized by

Google

124 Report op Committee on Reinforced Concrete.

Table X. — Continued.

Set

Load

Gauge

Point

204

205

200

207

208

209

210

2
DL

0.4

750

0.1

190

0.4

750

-0.8

-1500

-0.2
-380

-0.1
-190

0.3
560

0.6
1120

0.6
1120,

0.7
1310

8 I 9 I 10
5 ' 6

2.9 9.4 9.7 10.4 11.5 11.7 12.31 12.0
5440 17620 18190 19500 21560 21940 23060 22500

0.9
1690

1.6
3000

-0.5
-WO

0.7
1310

0.8
1500

-0.1, -^.3 -0.3
-190 -560 -580

-0.3
-560

0.9
1690

0.1
1880

3.11 7.7
5810 14440

0.1
190

3.2

5620

-0.3
-560

5
2810

0.6
1120

12.2
22880

1.7 .
3190|.

3.3
6190

4.7 5.2 4.0
8810' 9750 7500

8.8! 10.4 12 2 12.4 14.4
16500 19500 22880 23250,27000

4.6

-0.5
-940

1.6
3000

4.9
9190

0.7
1310

0.5
940

0.7|
1310

6.4 5.6 6.7
12000 10500 12560

0.3 1.0 0.6 1.1
560 1880 1120 2060

-0.3
-560

2.9; 3.1
5440; 5810

4.2 4.4 4 5
7780 8250 8440

7.0 6.3i
13120 11810

1.9 .
3560 .

5.0 .

15
10

16
DL

9.7; 7.1 5.80

18190 13310, 9380

2.3 2.3I 1.3

4310 4310 2440

8.0
15000

10.61 3.2
19880 17440

-0.8
-1500

4.8 3.2

9000! 6000

0.6| 0.9

1120. 1690

3.2 4.2 -2.9
6000. 2810-5440

1.4
3190

0.4
750

-0.8
-1500

Table XI. — Deformations in Inches and Concrete Stresses in
Lb. per Sq. In.

Set

Loadj DL

Gauge
Point
301

302
803
304
305
306
307
JC8
309
810

-0.8
-100

1 I 2

-0.3
-40

-0.7:

0.8 -0.1
100 -10

-0.5
-60

0.1
10,

-0.2'
-25

-0.2
-25

0.5 -0.7 .
60 -90 .

-1501.

0.1 .
10 .

-0.2
-25

-0.2;
-25

-0.2
-25

-0.7
-90

-0.6
-76

0.6.
75,.

-0.9'.

-no .

-1.61
-200

-1.0
-125

0.1
10

-0.8
-100

-0.7
-90

-2.8
-350

-0.3
-40

-0.9
-110

-0.7
-90

-08

-:oo

1.5
190

-03
-40

-0.7
-90

-2.3

-6.8
-850

-1.4
-175

-3.9
-490

0.3
40

-1.1
-140

-0 1
-10

11 12

-1.2 .
-160 .

-1.5 .
-190 .

-2.0
-250

-1.8
-225

-1.2

0.3;
40

,^.,1.

.1 -0.3 -0 2
i -40, -25,

-1.7 -2.6

-210 -325 -360: -425

-0.1 -Oil
-10 -10.

-2.9 -3.4

-3.1
-390

0.3

0.2
25

0.4 .
50 .

16
DL

-0.2
-26

0.2
25

0.5
60

-2.9 -0.7i -1.6
-360 -90 -190

0.1
10

-3.2-3.9! -.7 -2.3

-400 -360 -340 -290

-6.8' -8 3 -8.6 -9.4 -8.5; -8.7
-850 -1040 -1075 -1175-1060-1090

-1.8
-225

-5.6
-700

-1.7
-210

-5.61
-700

-1.4 -1.7.
-175 -2IO1.

-6.3 -7 9
-790 -990

-7.1 .

. -0 1 -0.4

.1 -10 1 -50

-7.1
-890

-6.2 -4.7
-775 -590

-0.1
-10

-5.1 -4 5 -2.7
-640 -560 -340

Digitized by

Google

Report of Committee on Reinforced Concrete. 125

Table XI. — Continued.

Set

1
DL

-0.7

4
1

10 1 11

Load

-0.4
-50

0.6
75

-0.6
-75

-0.4
-50

-0.4

-0.1
-10

0.4
50

-«.5
-60

-0.1
-10

-0.2
-25

0.5
60

-0.6
-75

0.5
60

-0.6
-75

0.1
10

2 1 3

4 5

1
i

Point

-0.6
-60

-0.8

-0.3
-40

-2.1
-260

-1.4
-175

-0.5
-60

-1.5
-190

-1.0
-125

-0.6

-90

-0.4

-50

-0.8

-100

-1.5

-75

312

1
-1ft

-2.9' -3.5
-3601 -440

-4.1' -5.5
-510 -690

-1.0

-lOO' -225

-135

313

1
-1.3, -1.6
-160 -200

-0.41 -0.8
-60| -100

-1.21 -1.9
-160 -240

-0.9 -10

-1.2

-150

314

0.1

-190

-1.1
-140

-0.6
-75

-0.1
-10

315
401

-1.3
-160

-10
-125

-3.1
-^90

-0.5
-60

-2.9 -3.6
-360 - 460

^).4'

-3.9
-490

-3 2
-400

-2.4
-300

-3.0
-375

-2.1
-260

-1.1

-110 -125 - -

-60

'..'.'.'.\..y.'.

-140

402

-0 6i -1 6

-2.4

-75| -200

-0.51 -1.0
-60| -125

-I.9I -3 2
-240 -400

-300

403

-0.1
-10

-1.0
-125

0.1
10

-0.5
-60

-0.2
-25

-0.8
-100

0.2
25

-0.2
-25

-0.1
-10

-0.2
-25

-0.5
-60

-1.0
-125

-0.9
-110

-1.1
-140

-1.4
-175

-4.5
-560

-2.3
-290

-3.3
-410

1
-2.4 -3.4
-300 -425

1
-2.6

-2.1
-260

-1.2
-150

-0.8

-100

404

-3 2

-325 ...

::::: ::.:.

-400

405

0.2
25

-1.1
-140

-0.1
-10

-1.4
-175

-0.2
-25

-2.6
-325

-0.2
-25

-3.0
-375

-0.9
-110

-2.4
-300

-0.3
-40

-3.1
-i390

-1 n

-1 8

-225

406

-3 3

-410

-0.5
-60

-3.4
-425

-2.9
-360

-0.7
-90

-3.5
-440

-3.1
-390

-0.9
-110

-3.8
-475

1
-2.9, -2.9
-360' -360

1
-0.8, -1.6

-no: -200

1
-3.7, -4.7

-460, -590

-3.2
-400

-2.7
-340

-2.4

-300

407

-1.6

-200

408
400

-4.7
^90

-3.6
-450

-2.6
-325

-1.9
-240

-1 5

-190

410

-0.5
-60

-0.7
-W

-0.6
-75

-0.3
-40

-0.2
-0.6

-2 5

-9. 7

-3.1

-390

-3.4 -3.6
-425 1 -450

-4 5

-1.0

-310; -340

-560

125

411

-1.1

' 1

::::: :::::

-140

412
413

-3.2' -3.3
-4001 -410

-1.5' -2.0
-1901 -250

'h).8

1 -100

-1.4' -1.4
-175| -175

-3.8' -3.4 -4.0
-4751 -I25I -.500

-2.6 -2.7, -3.3
-325' -340l -410

-1.3 -0.9, -0.8
-160' -110| -100

-1.9' -2.4| -3.1
-240| -3001 -390

-1.4I

-650|

-3.8
-475

-2.4
-300

-1.6
-200

-1.6

-75[ -125
-0.5' -0 8

-200

0.2
25

-1.6
-200

0.1
10

414

-60

-0.4
-50

-0.1

-100

-0.7
-90

-0 ?

415
417

-0.2
-25

-0.1

-0.7

-3.2
-400

-2.6
-325

-1.7
-210

-"1

-10 -25

i""[

.........j. ... ..........

Digitized by

Google

126 Report of Committee on Reinforced Concrete.

— — d MM y

Un/t Tension in Jteei

400*

)

=*, 1

{

00 5000

[»7 ,

5000

ifd 1

5000

\or

1 5000

Unit Tension in sreel

^ 400
^JOO
5 ?00
I /OO
S

p I J -i t

~- t X t

r^ ^ 4 - t

ST
^500

5 ?«>

/OO

•^

107

loe I09

ikW"

///

unit 5fre55 in 5tee\

115

114

\II5

-TSSr

^400

^00^
500

1 /a?

//7

(' 3000

116.

J-

IZO

-L

6000 MOO

Unit 3tre3S in 3teel
FIQ. 51. — LOAD-DEFORMATION DIAGRAMS FOR REINFORCEMENT.

Digitized by

Google

Report of Committee on Reinforced Concrete. 127

o 3600 6 foooo

Unit Tension in 3feel

^400

I /oo/-

ZIO

-1-

ZIZ

6^ tOtOOO b MXXX) O 5000

Unit Tension in 3teel

?I6

O to 20 30 '40

Distonce from Capital in Inches

^400

2 Joo
I o

1 — ^\

i^zta

(

6 A

Unii

^ 3tr

ess

in Jj

^eel

IM I

) t

}6t

tdi

\dd

Unit Stress in Steel

PIG. 62. — LOAD-DEFORMATION DIAGRAMS FOR REINFORCEMENT.
READINGS AND CURVES BY W. A. SLATER.

Digitized by

Google

128 Report op Committee on Reinforced Concrete.

to trace, and were only of such magnitude as would be expected
to accompany the reinforcement stresses observed.

^500
S zoo

Unit Compression //? Concrete

too

I \

9 .,/

3/s

» c

) 5

00 i

■> A

00 c

30 C

) 300 C

> 600

Unit Compression m Concrete

Unit Compression in Concrete

Fia. 53. — ^LOAD-DEFORMATION DIAGRAMS FOR CONCRETE.

Examination of the plotted deflections and stresses shows a
change of inclination of the curves at a load of 200 lb. per sq. ft.
Take, for instance, point 202, which is typical of this condition.

Digitized by

Google

Repobt of Committeb on Reinfobced Concrete, 129

At 200 lb. per sq. ft. the stress in this bar was 14,000 lb. per sq. in.;
the load producing this stress was that of stage 3.

Unit Compression in Concrete

4aL

fam loos

JUX)
; 300

Ebb 6 Ad 6 ido 5
Unit Compression In Concrete

407 4408

So-

J40i \

o soo

I ^^f—l — T

:^ \40§ \4IC

« % lo h A

\4II

300

6 SCO o 355 soo o soo'
unit Compression in Concrete

1 4/3 \4K \4/5\
> soo 500 o soo'

FIQ. 54. — LOAO-DBFOBMATION DIAGRAMS FOR CONCBBTB. RBAAINQ8
AND CURVES BT W. A. SLATER.

When the load from the adjoining panels was shifted and cor-
responded to 400 lb. per sq. ft., the stress was only 19,000 lb. per
sq. in., instead of some 32,000 obtained by producing the curve to

Digitized by

Google

130 Report of Committee on Reinforced Concrete.

the 400 lb. line at that slope which obtained below. This indicates
that a part of the load is carried by the panels adjoining the loaded
one.

The same result is indicated in the case of point 104, which
is at the bottom of the slab near the middle of one of the panels.
The fact that the adjoining panels assist in carrying a loiad placed

Table XII. — Summary op Stresses.

Location.

Gauge
Point.

Design Load,
200 lb. per sq. ft.

400
lb. per«q. ft.

D.L.*

L. L.

Total.

L. L.

Total.

Reinforcement Streui
Over Column Head

Long Span

202
204
Av.

2.000
2.500
2.250

14.000
17.500
15.750

16.000
20.000
18.000

19.000
23.000
21.000

21,000
25.600
23,260

Short Span

201
206
Av.

1.500
1,000
1.250

3.000
3.000
3.000

4.500
4.000
4.250

7,000
1,000
8,600

8,600
11.000
9,750

Bottom of Slab at Side of
Panel

Long Span

102
112
Av.

1.000
1.600
1.250

7.000
11.500
0.250

8.000
13.000
10.600

9.000
13.600
11.260

10,000
15.000
12.500

Short Span

101

500

1.000

1.500

4.000

4.600

Bottom of Slab at Middle of

Long Span

104
122
Av.

500

1.000

750

7.500
4.000
6.750

8.000
6,000
6.500

12.000
9.600
10.760

12.600
10.600
11.600

Panel

Concrete Streaees,
Bottom of Slab at Colxmm

Short Span
Long Span

103
109
Av.

306

500

1,000

750

100

1.500
2,500
2.000

750

2.000
3,500
2,750

850

8.000
6.600
7.260

1.060

8.500
7.500
8.000

1.160

Short Span

315

100

250

350

450

660

Top of Slab at Edge of Panel

Long Span

406
412
Av.

100
100
100

300
400
350

400
500
460

360
600
426

460
600
625

Top at Middle

408

150

400

660

600

750

Bottom of SUb at Wall

Long Span

302
304
Av.

50
100
75

150
250
200

200
350
275

260
400
326

300
600
400

* Dead-load Btreases obtained from atreas-deformation curves by projecting curve down to
no-load line.

on one panel is also shown by the fact that deflections were ob-
served in panels adjacent to those loaded^ as shown in Fig. 48
and by the fact that cracks on the under side of the slab were
traceable well past the middle of panels adjoining these loads.

The point of contraflection was foimd to be at 0.21 of the
clear span from the edge of the column head.

Digitized by

Google

Report op Committee on Reinforced Concrete. 131

FIG. 55. — PLAN SHOWING LOCATION OF CRACKS. FULL LINES INDICATE CRACKS
ON TOP AND DOTTED LINES ON BOTTOM OF SLAB.

Digitized by

Google

132 Repobt op Committeb on Reinfobced Concrete.

. Mm

FIQ. 66. — BARR BUILDINQ TEST PANEL, ST. LOUIS, MO.

Digitized by

Google

Repobt op Committeb on Reinforced Concbete. 133

The high values of bond stress observed indicate that con-
sideration be given to this feature in the design of flat slabs,
especially when it is realized that no matter what scheme of rein-
forcement be adopted there will exist stresses at right angles to
any given bar over the column which tend to destroy the adhesion
of the concrete to the reinforcement.

Part III. Test op a Concrete Floor Reinforced in Two

Directions.

Test Panel — ^The test was made on a panel representing a wall
panel of a continuous floor and to approximate existing conditions,
cantilevers were built on 3 sides. These cantilevers were loaded

FIG. 57. — BARB BUILDINQ TEST PANEL WITH LOAD OF 380 LB. PER SQ. FT.

during the test until the tangents to the slab over the supporting
beams became approximately horizontal, as would be the case in a
continuous floor under multiple panel loading.

The panel proper is 25 ft. long by 26 ft. 9 in. wide, Fig. 56;
and is carried by steel I beams, fireproofed with concrete. These
beams rest on steel bearing plates on concrete posts at each comer
of the panel. Seven-inch tile was used with a cover of 2J in. of
1:1^:3 concrete, making the total thickness of the slab 9 J in.

The arrangement of tile is such that when alternate rows of
block-tile and channel-tile are laid, the concrete is formed into
two series of intersecting T beams, spaced 15 in. centers each way
and having 3-in. steins. Small furring-tile, laid at the intersection

Digitized by

Google

134 Repobt op Committbb on Rbinporcbd Concrete.

of the ribs thus formed, complete the tile ceiUng surface. The
panel was remforced as shown m Fig. 56, designed for a live load
of 150 lb. per sq. ft.

The panel was poured on September 1, 1911, imder unfavorable
conditions, it being necessary to carry on a considerable portion

PIQ. 58. — PLAN SHOWING METHOD OF LOADING BARR BUILDING TEST PANEL.

of the work after dark. The weather was very hot during the
next few days and several of them were holidays, so that the slab
was not even kept wet. On the whole, conditions under which
this slab was erected were probably not more favorable than would
exist in an actual building.

Digitized by

Google

Report of Committee on Reinforced Concrete. 135

Method of Testing. — Sand in sacks was used for the load,
Fig. 57, each sack being filled on the scales to weigh 100 lb. The
loading was started on November 23, 1911, and progressed by
stages, alternating with readings, until 650 lb. per sq. ft. was
reached on December 6, 1911. The test was stopped at this
point because of the expense involved in placing the load upon
the high piles of sacks. The total load remained in place on the
panel for 48 hours, after which unloading was started and completed
5 days later.

Table XIII. — Stages op Loading.

Time of Loading.

Load in lb

periQ.

ft.

Stefe.

Gantileve

Start.

Finiflh.

Panel.

Left.

Rear.

Rilht.

1911

Nov.

23. 2.30 p.m.

Nov.

24, 10.00 A. M.

75.5

38.3

38.6

39.4

Nov.

25, 11.00 A. M.

Nov.

25. 3.30 p.m.

76.5

74,3

70.3

74.3

Nov.

27, 2.00 p. M.

Nov.

27. 4.00 p.m.

75.5

98.2

Nov.

28. 12.30 p. M.

Nov.

28. 6.30 p.m.

150

183.5

194.2

191.5

Nov.

29. 12.30 p. M.

Nov.

29. 5.20 p.m.

247

183.5

194.2

191.5

Nov.

30. 1.00 p.m.

Nov.

30. 4.50 p.m.

300

183.6

247

191.5

Dec.

1. 10.00 A. M.

Dec.

1. 2.10 p.m.

300

298

300

Dec.

1. 4.30 p.m.

Dec.

1, 6.15 p.m.

300

346

298

346

Deo.

2. 9.30 a.m.

Dec.

2. 9.50 a.m.

300

372

298

372

Dec.

2, 10.00 A. M.

Dec.

2. 10.15 A. M.

300

385

298

385

Dec.

2. 12.05 p. M.

300

385

298

385

Dec.

3, 9.00 a.m.

Dec.

3. 9.45 a.m.

300

385

298

385

Dec.

3. 9.00 a.m.

Dec.

3. 2.00 p.m.

380

465

376

465

Dec.

5, 10.30 A. M.

Dec.

5. 4.00 p.m.

500

465

376

465

Dec.

6. 9.40 a.m.

Dec.

6. 10.40 A. M.

527

465

376

465

Dec.

6. 11.20 A. M.

Dec.

6. 12.15 p. M.

550

465

376

465

Dec.

6. 12.50 p. M.

Dec.

6. 1.35 p.m.

567

465

376

465

Dec

6, 2.10 p.m.

Dec.

6. 3.00 p.m.

591

465

376

465

Dec.

7, 8.00 a.m.

Dec.

7. 9.20 a.m.

615

465

376

465

Dec.

7. 10.60 A. M.

Deo.

7. 1.10 p.m.

650

465

376

465

Both reinforcement and concrete stresses, as well as deflec-
tions, were measured throughout the test, exceptionally complete
sets of readings being taken imder the design load of 150 lb. per
sq. ft. and under the test load required by the City of St. Louis,
t. e,, a superimposed load equal to once the dead load plus twice
the live load, or 380 lb. per sq. ft.

Deflections were read by measuring the distance between a
steel plate fastened to the ceiling and a steel rod held in a scaffold
below. An inside micrometer reading to 0.001 in. was used, both
plate and rod having countersunk holes to locate exactly the posi-
tion of the instrument.

Digitized by

Google

136 Report of Committee on Reinforced Concrete.

The stresses in the remforcement were read by an exten-
someter which is a modification of the Berry Strain Gauge used to
read stresses on the concrete, while W. A. Slater read reinforcement
stresses with a modification of the Berry Strain-Gauge, made at

FIG. 59. — PLAN SHOWING LOCATION OP DEFLECTION POINTS.

the University of Illinois, Figs. 1 and 2. Both of these instruments
have points which are inserted into small holes drilled in the steel
or in metal plugs set in the concrete and read directly to 1/5000
of an inch, or by estimation to one quarter of this amount. An

Digitized by

Google

Report op Committee on Reinforced Concrete. 137

Table XIV.-

-Deflections, Barb Building Test Panel.

Unit
Load.

76.5; 150

300

380

380 500

500

627

550

567

591

615

660

Deflec

tiOD

Pomt

Stage.

Date
and
Hoar.

•^1

• si

.354
.266
.198
.153
.130
.117
.110
.127
.142
.134
.296
.213
.160
.104
.094
.089
.078
.086
.090
.091
.071
.054
.043
.034
.024
.019
.022
.025
.030
.116
.098
.082
.075
.068
.073
.068
.072
.082
071
.063
.054
049

".Hi

.903
.724
.559
.428
.420
.421

1.340 1.417!l.456
1.043{1.099ii 2lfl

1.456
1.216
1.017
.870
.952
1.167
1.332
1.631
1.921
2.155

.794 831

.591
.567

.611

.581
.568
.658
.778
ROT

.480; .645
.568' 764

.645
.693
.860
.598
.438
.302
.317
.310
.352
.408
.460
.067
.402
.331
.265
.205
.172
.148
.154
.165
.179
.530
.448
.362
.293
.262
.244
.238
.204
.248
.386
.324
.260
.202
.177
.159
.167
.186
.202

.878

.9601 .982
1 153i 1.222
.890' .940
.6361 .668
.419, .438
.424) .434
.4101 .417
.468, .478
.5461 .558
.622| .633
.073 .078
.5441 .553
.4441 .453
.354 .358
.268 .274
.234 .247
.228' .239
.2711 .290
.330, .355
.3941 .431
.7251 .742
603 616

19
21

Col.

.02

.100 .103

.107

.109

.114

.126

.137

'28

.477
.376
.351
.351
.373
.415
.467
.527
.432
.338
.251
.228
.232
.263
.324
.880
.066
.155
.900

.488
.384
.361
.366,
.3921
.442
.502
.545
.446
.347
.258
.238
.2451
.285,
.3501
.414
.068
.1581

02fi

I"":

046

.053
.064
.079
.095
.016
.042
.104
.274
.124
.229
.013
.071

!!.!. :....

48
49
60

Col.

.003
.027

.033' .047
.118, .119

' .643

.7741 .925

.72.'5

.085
.264

.087 .094 .095
.2741 .2941 .324

■ 1-

.099
.348

.103
.379

.107
.414

.115
.409

51
52

Center

.129

1.29911.331,1.947
.993 1 0171

2.6192. 163'2. 327

1... 1. .

2.444

2.658

2.897

3.313

53
54
55

Col.

"cor*

.009
.040
.039

.152
.211
.114

.127
.335
.078

.148
.434
.105

.154
.449'

.113|

.162
.535
.130

.175 .176, .171
.5451 .56ll .586
.136| .139 .143

.170
.601
.147

.175
.627
.150

.190
.659
.158

.196
.702
.165

:::::i:::::

Corr

0.080

.167

.594

.641

.920

.9441.459

l.sie

1.632

1.777

1.872

2.052

2.252

2.608

Laat line of readings are deflections at center of panel assuming supporting beams to be rigid. All
other readingB eorrected for pier settlement only.

Digitized by

Google

138 Report op Combiittee on Reinforced Concrete.

8 in. gauge length was used throughout the test. The length
of such an instrument varies with changes of the temperature of
the observer's hands, to correct for which readings were taken at
intervals on standard bars and all temperature corrections neces-
sary have been applied to the stresses given in the tables and
plotted in the curves.

Method of Loading. — To prevent the possibility of the load
arching, the sacks of sand were arranged in separate piles, as

&50

^^^

,^'

--"

J>^'

A/)/)

0\iftlO

^JMD

/mi

L_&

p.SL±U

rstLoad

a«^^c

|5<»

Curve / 5ha¥5 defk

ortlon of h

r -

siqn

Lope

center of the panel wilt) reelect to
the columns.

Curve t shows deflection of
Jhe center of the panel with respect

S/oo

>«4

to the supporting beams

. 1 . 1 1 1

^Xl.

ilea

4 d.

6 d.
in

6 /
/nch

ea.

FIG. 60. — DEFLECTIONS AT CENTER OF PANEL.

shown in Fig. 57. The wide aisles, Fig. 58, were necessary for
the accommodation of the instruments and observers. Table
XIII gives the loading on the various portions of the panel at
different times, as well as the loads per square foot of slab and
cantilevers and the date and hour at which each stage was started
and completed. The loads per square foot are in each case
obtained by dividing the total load on the panel or cantilever by
the area, the intensity of load under the piers being much greater.
Up to the test load required by the City of St. Louis the

Digitized by

Google

Repobt of Committee on Reinforced Concrete. 139

Deflection Points

QJS^XJfM^XJl JO

Deflection Curve along Center- line of Panel in Direction of long Span
Under Design load {I50*per x^n)
Deflection Points

fl J

«J

7 S

6 S

5 34 33 S

r s

3t 34

did

K ^

^ \

X #

^ 1

•^

V w,

/ !^

1-0

>v J?

y «

I.I

^s.

>^k 1

/«

Deflection Cunn along Center- line cf Panel In DIrectloh of long Joan
under Test Load required by the 3t Louis DIdg Code (5acrperjif.fr.)

WIQ. 61. — ^DEFLECTION CXTRYE IN DIRECTION OF LONG SPAN.

1 i

ilea

Pomtj

3 i

r f

10 a/ s

4«

^ — 1

-r

^ 1

.50

flection

rve

a/a
una

\iisign lSoS (lOL

^;jr"4^'^'

Tf Short span

1 J

)efk

vtk)

nPi

vnt.

5 7 i

> s

} I

.to

—

M^l

•k

■^

^ \

•4g

/ \

**»

LtO

/JO

Defkchon Curve along Center- line of f^jnet in Dined fop of3hort Span
under Test load required fy the St. Louis 3/dg Code(3d0^prs(f.ft)

FIQ. 62. — ^DEFLECTION CURVE IN DIRECTION OF SHORT SPAN.

Digitized by

Google

140 ReFOBT of ComUTTEB ON ReINFORCEO Ck)NCRIiTB.

cantilevers were carefully loaded to horizontality, but from this
stage on no more load was applied to the cantUevers and they
rose slightly, thus increasmg the deflection at the center of the
panel.

I--!- Kr-««"B55— B» ^

gHDnnnnnniiBannnnDD
TinDDDnnnnnnnnDnnnn
DnDnnDDnnDnnnDDDnd
DnDnnDnDDDnnnnnDn
nnnnnDDnDDnnnnDDDD
nnDnnnnDDDnnnnDnnn
DDDDnDnnDDnnDnnDD
nnnnnnnDDDnnDDnnaD
DnnDnDDDDDnDnDnDD
nnannnDDDDDnnDnDnL

•-• Indicates Points onfop of Slab.
o-o Indicates faints onBoltomafSab

aDDDDDDaDDDDDnnDa

nnnnnnnnnnDnnnnnn
DaDnanDDnnnnnnnDD .,
DDanDDnnnnDnnDDnnngj
pnnnnnDnnDnDannnnn^
DDnannnnnnnnnDnnna

DDDDDDDDQDDDDDDDD

^nDnannnnaDDnnnnnn

no. 63. — PLAN SHOWINQ LOCATION OP GAUGE POINTS ON CONCRETE.

The horizontality of the tangents over the beams was deter-
mined by measuring deflections of both the slab and cantilevers
at a series of points along a line perpendicular to the beams.
The deflections were then plotted and a smooth curve drawn

Digitized by

Google

Report of Committee on Reinforced Concrete. 141

which showed at a glance whether or not horizontality existed.
It was often found necessary to shift the load on the cantilevers
several times before the correct amount and position were
reached.

.40)

4O0

b 500 Kxx> o mTIoSS ^ ,
Unit Compression in Concrete

SCO 1000 O JoO /OOO 1500

tddd d 3^ md o S^ }6do

Point5ii

O 300 1000 O 500

Unit Compression In Cortcnefe

FIG. 64. — ^DIAGRAM OF STRESSES IN CONCRETE.

:305-

Deflections. — Deflections were read at the points shown in
Rg. 69. A summary of the observations is given in Table XIV.
There was a settlement of the footings which has been corrected
for in the summary and in the curves.

Digitized by

Google

142 Report of Committee on Reinforced Concrete.

The center point deflections are plotted in Fig. 60. The
deflections shown by the curves represent the deflections of the
panel with respect to the supporting beams. In order to arrive
at the deflection of the panel proper, the gross deflections were
decreased by an amount equal to the average deflection of

3600

2400

2200

ZOOO

%I600

I f600

^ I400

^l2O0

%I000

. 600

§ 600

% 400

ZOO
O

^•*"

a-*"

,^^

hroke

,000£ J00O4 .0006 .OOOS

Unit Deformations '(6 gauge length)

.ooto

PIG. 65. — STRESS-STRAIN DIAGRAM OF TWO CONCRETE TEST CYLINDERS.

the supporting beams. Curve 1 shows the gross deflections,
the dotted curve including settlements occurring during the
intervals between loads and the solid line giving the true elastic
deflections (including beam deflections) caused by a uniform rate
of loading. The dotted portion of Curve 2 ehows the same thing

Digitized by

Google

Report of Committee on Reinforced Concrete. 143

after correcting to a uniform rate of loading. The solid line,
Curve 2, should be used in comparing this test with others or as
a basis for deflection coefficients.

T^-^-fi

-f

' DnnnDnnb"

otpt09

-"IBS-

DDDnnnn

,nnnnDnna

DDDDnnan

.nnanannn

nnnnnaD

«r »* M

DDDDDnnnis

DDDnaaDa
nnnnnnnnjs

DDnDDnaDj

nnnDDDDnT

DDDnnnnnraannnnDDni^

nnnnnnanDaDnnnnnnDT
aDnnpnaDDDDDn '"

]0[

nnnnnDnnDnnnnDnDDD

DnnnDnnnanDnnDnDDD

DDnnnnDDDDDnnDnnDn

nnnnnnnDnDDDDnnDaa

nDDDnnnnDDnnnDDnnn

DnDnDnnnnDDDDDnnDn

DDDnDnnDDDDDDDaann

nnnnnnnDDDDDnnaDnn
^ ^j

•-• Indicates Points on Top of Slab
oo Indicates Points on bdttom of5lab .

PIQ. 66. — PLAN SHOWING LOCATION OP GAUGE POINTS ON REINPORCEMSNT.

Fig. 61 shows exaggerated scale sections of the panel along
the long axis for the design load and for the load specified by the
St. Louis Building Laws. The deflections, from which these are
plotted, are corrected for settlement of the footings only. Fig. 62

Digitized by

Google

144 Report op Coboiittee on Reinforced Concrete.

shows similar sections along the short axis of the panel. Inspec-
tion of these figures will show that the cantilevers were properly
loaded to obtain horizontal tangents over the supporting beams.
Deformations and Stresses. — Concrete stresses were read at-
the points shown in Fig. 63. A summary of the observed stresses
is given in Table XV; the loadnstress curves at a few of these
points are plotted in Fig. 64. In these curves and in the sum-
mary, the stresses are those which were induced in the concrete
by the loads greater than 75 lb. per sq. ft. At this stage the
points, at which these stresses were read, worked loose and had
to be reset.

Table XV.— Concrete Stresses, Barr Building Test Panel.

Unit
Load.

150

160

800

380

880

Gaufce
Point

Stage.

Date

Nor. 28

Nov. 29

Dec. 2

Dec. 3

Dec. 4

and Hour.

7 p.m.

10 A.M.

12 m.

8.30 p. M.

9.30 p.m.

401

R.O.

-40

-285

-81

-625

-796

-765

402

R.O.

+40

-162

+ 122

-430

-422

801

R-0.

—195

—203

—943

—1225

—1322

803

R.O.

-195

-504

-211

-1005

-1322

-1370

805

R.O.

-105

-350

-57

-740

-935

-1022

802

R.O.

-26

-858

-57

-723

-870

-967

804

R.O.

-122

-382

-162

-796

-1022

-1095

808

R.O.

-58

-252

-73

-1005

-862

-926

810

R.O.

+65

-219

+130

-740

-796

-943

807

R.O.

+146

-471

+90

+130

+365

+211

809

R.O.

-49

-398

-81

-292

-220

-374

811

R.O.

-80

-325

-560

-480

-600

+ as Tension.
— »Compreimon.

The reduction in stress with increasing load appearing in
some of the curves may be the result of the gradual giving away
of the concrete in tension and resulting shifting of the point of
inflection.

Two test cylinders 8 x 15 in. were taken from the mixer at
the time the slab was poured and after being cured under the
same conditions as the panel itself, they were tested just after
completion of the readings on the panel. The ultimate strength
and the modulus of elasticity of these were widely different as
can be seen from the stress-strain diagrams in Fig. 65. In deriv-
ing the concrete stresses the average modulus was used, being
taken at 3,250,000.

Digitized by

Google

Report of Committee on Reinforced Concrete. 145

*<»9i

"si

ecrt
4 +

II
++

|i"

I
I

II II ll II

e5« SS 5w e«e«

++ +4- +4-

§1 i§ §1

<DO C4C4 OOOO

^H«H <^<^ ^4<-l

++ +4- 4-+

4-4-

II II

4-4- .4-+

11 ii
-f4- -f4-

+
8

r^ -

§1 11 ii 11 i|

Soi eoeo «c8 «w «eo

4-+ -f+ -f+ -f+ -f-h

§i §1 §§ i§ §§

sa S2 sa 8"s a:s

4-+ 4-4- 4-+ 4-4- 4-+

§;i;i§ii;Siil§

I I

4-1

Digitized by

Google

146 Report of CoMHiiTTEB on Reinforced Concrete.

<
fit

+35W

: :'

i i

i Hi?

! :1

f :

: 4

i +? ?$ f ? j!

: ++ i+ :+ :■

: +

i §

: +

: 1
: o»

: +

i t

i i

: +

III

i ^ II II 1! I| i!

: ++ ++ ++ ++ ++ :■

1 p ti ^

i ^s sS ?$
^ ++ ++ +^

^ J

«^^

11 i

^S 3
++ +

i |§ ig 11 IS IS il m II il

i ^i i+ i+ i+ T+ n s2 == %%

- 1

•«

: S

\ i
: 1

; §

: +

■ +

: +

; +

: g

: +

i g

•*

; +

: i

: S

i t

•O

: 1

: o

\ ?

: 1

1 1

1 7

: +

j +

: +

1 •

n n
+ 1

Digitized by

Google

Rbpobt of CoMMiTTBfi ON Reinfobcbd Concbets. 147

3 ii

r f

f :■

18!

i if

l It

5 i

r :

f :■

<
Oh

; 8

: !J

: +

i — :-

:— :

<
H

i 1

i i

: 4-

g
S
g

ii 11 II II 1
?^ 1? ?? $^ +

1^1

i 11 II ll If

? 77 77 ++ +'

*<•

nil

4- ++ +

i 11 i| 11 m \

III
T 77

i S

: 1

; +

. §

i §

: ::-

: +

: + :

i 1

: +

; 1

1 ^

': +

§ § s

I I

+ 1

Digitized by

Google

148 Report op Committee on Reinforced Concrete.

Mi"

11^

ii"

11 II II

tz ++ %%

i| §1

++ ++

i| §1 11 II II II i

Vh «4- ooT ■^co ■"♦"o
I -I. ' ri-i- -^r" -^-^

+ 1 +

■*« '♦O c«

++ ++

I I

04 04
i I

i§ i§ §§

I I

T§'

: i

" : 1

; 1

: o

t>-

• : +

-i— ^

—

-:- -:

s s s ss s;

e4 *i4 1-4 v^

II n

+ 1

Digitized by

Google

Report op Committee on Reinforced Concrete. 149

-•—

1 \

i i

s ;

+ :

1 i

4- :

1 :

4- :

? 1

1 ^

1 \

? :

-:-4-

-; -[

-:— f

i--i-

»«

i.'

i~F

— :

\ — :

4—^

-:-

<<l

§ ii 11 1

§ §1

1 s§ §!

f +4- +

"t" t* k *

1 §§ §? §<
+ :f+ + +

7+ :

CI .

i +

: a

4- 'r

i g

i 1

: +

: J2

-i—

'i*

-f

-i-f

HO
I
+ 1

Digitized by

Google

150 Report op Committee on Reinforced Concrete.

i 1
; S3

: -t

i 1

1 ^

i §

: or
: +

: +

««•

: 1

: +

ill

eo<4< CO
1 1 1

i §
i ^

•*

SB-"

i U

: +

111

!•

HO
H I
+ 1

Digitized by

Google

Report op Committee on Reinforced Concrete. 151

Fig. 66 shows the location of gauge points at which reinforce-
ment stresses were read. A summary of the observed stresses is
given in Table XVI, together with the load and time at which
each set was taken. The two figures given for each point are
by the two different observers. Figs. 67 and 68 show loadnstress

700

--•

Xldr^

.^^Siid

• Simmi

JSW

oint

201

7/nf

?«?

De^

^nu

•

200

(

toe

^00

Unit Tension in Steel

700

^600

—

-/"/fw

, j-^ «i

tfutim

n^ -

ZySlJiLi^SMaW.

TSL.

'"'

■"

Point 219

■>lnt

113

—

.^S&ML^

U)ad

—

{

(

nit Ti

100

vrsiOi

40C

1 In

100

Jfee

, '

)

tXK

XX)

40(y

no. 67. — ^DIAGRAM OF STRESSES IN REINFORCEMENT.

curves for a few of the bars at critical points and for the struc-
tural steel supporting beams.

Fig. 69 shows the distribution of stress in the bars which are
on the short span. The section shown is taken along the long
axis of the panel and the stresses plotted are those in the bars
cut by the section. Similar curves are shown in Fig. 70 for the

Digitized by

Google

152 Repobt of Committee on Reinforced Concrete.
TOO
^600

\600

JOO

Ux)di

-r-— -

^ 1 1
Point IZZ

Point I2Q

1 1

—

,_D^lgn_Loa<L- _

— .

—

O tOOOO 40000

Unit Vuision in Steel

gOOOO 40000

^600

poo

1»9

sax?

100

Point ioi-\
Outer Beam jr

/ /" Point 103
( /Inner Beam

Point 124
Outer Beam

Iru^A

^ Point 126
Inner Bean

~""~

bLJdnLfL:

^t.UJU

±!S^

ipga_

■^~"~"

¥'-

■

/ Points 101 A

105

/Points lZ4d

l?6

—

_j?

?^£'7

Load

.„

—

{

5-^

■ tOi

odd

4o^

000

)

zoo

)00

40i

000

Unit Tension in Steel

FIQ. 68. — DIAGRAM OF STRESSES IN REINFORCEMENT AND IN STEEL
SUPPORTINQ BEAMS.

\40000-
% 30000

\toooo

\ 10000
O

no. 69. — ^DISTRIBUTION OF STRESS IN THE REINFORCEMENT CABBTING THE
LOAD ACROSS THE SHORT SPAN.

Digitized by

Google

Report of CoMinTTEE on Reinforced Concrete. 153

top bars over the supporting beams. Some of these curves may
be distorted because the supporting beams tipped up on the
bearing plates and spread apart at the comers, thus inducing
tension across the diagonals of the slab.

V. ' . *".4. ' . ' . ; .".'- ' ^ » .v.m.nij r »...^.ni. i i.-;.-.i ' .i.^\-.n-'. . ■ .,*■•■■..>.• ... I ... ■.. ' •'••• .'..S-'V ^» '.*.■. ' ■• i-. ■ ■ ' • ' •1 *1

Dbtri button cf Jtress In Top dars o/er the Z5'cf Supporting deam

f ^ ■•■■ • :■' v-:^|. ' •::;:• ^^•■^^:v;^:^^^T:^.^^•^^•■:•^'^^^*••V••?|• ^' •"•^

P0intZ09 Z07 Z05 Z05 ZOI

Distribution of Stress in Top Dars arer d6'9*
Supporting Beam- Dars extend into Cant Her er

10000

ointzn z

\SQO

}\//in^

hi30

Point Zn ZI5 Z15 ZI7 ZI9
Distribution of Stress in Top 5arj
over Z6''9'' Spandrel Ream

FIQ. 70. — DIAGRAMS SHOWING DISTRIBUTION OF REINFORCEMENT STRESS.

At the 500-lb. load, cracks could be found extending a short
distance from the comers of the panel along the diagonals.
Pig. 71 shows the location of cracks at a load of 660 lb. per sq. ft.

Digitized by

Google

154 Report op Committeb on Reinforced Concrete.

In order to locate the points of inflection and to determine
the bond stresses; readings were taken along portions of three
bottom bars. The results of these readings are plotted in Figs.
72 and 73.

^'1 1 1 1

u 1 1 1 1

I I

I
I
I

annnnDannDDaai

gaDDaaaaDannnDQa

nnnnannnDaanDaaa

I cgoannsaaDDDDaDnaaa
I nannnnanDDDDnDDaan
I nnDDDDannDBoonannn

f-SDnnnnnDnDanaannna

D
D
D

aaanDaDnannaDnDDDD

nnnaDDDaQDnnanna

DDDn

DnnaQBaDE3QaDDnDD
nnnnnDDnnaDaaaDD
DnDnDDnnannnnnDa

□DC

DDD[

DDDDD

! DaDDanDannnannaD!

aDDDDDDDnaDDDDnn
DDDDDDDaDDDDa

DDDDDDannnbg

3 QEPDDaDDDC

JDDDC

FIG. 71. — PLAN SHOWING LOCATION OP CRACKS.

A summary of the stresses and deflections at the critical
points of the slab is given in Table XVII.

No definite conclusions as to the distribution of stress among

Digitized by

Google

Report of Cobimittee on Reinforced Concrete. 155

10000 §*^

zoooo ^^

a^^imyffmfM

^ o

^^zoood

Point of inflection 30Cf '

Load-/30*.
tjoacl-300'

Load-36(f

no. 72. — VARIATION OF STRESS AND LOCATION OF POINT OF INFLECTION ALONO
MIDDLE BAR IN DIRECTION OF SHORT SPAN.

Digitized by

Google

156 Report of Committee on Reinforced Concrete.

Table XVII. — Summary of Principal Stresses and Deflections, Barr
Building Test Panel.

Deflections at Cbnteb.

Design Load 160 lb. per sq. ft.

St. Louis Building Code 380 lb. per sq. ft.

0.167 in., or j^„ of Spian

0.860 in., or gf, of Span

CoNCBBTB Stresses.

[>de
l.ft.

Gauge
Point.

Design T.oad
160 lb. per sq. ft.

St. Louis C
380 lb. per sc

L. L.

D.L.

Total.

L.L.

D. L.

Total.

Top of Slab at
Center

Long Span
Short Span

402
401

100
100

200
200

675
800

100
100

775
900

Bottom of Slab at

Flange of
Supporting Beam

Lon

Short
Span

*1Ci
End
Free
End

304
801

300

325
400

160

150
200

475
600

225

1,250
1,525

425

150
200

1.400
1.725

600

Bottom of Slab on
Rib just outside
of Beam Flange

LOD

Short
Span

End
Free
End

308
303

311

175
400

175

75
200

100

260
600

276

926
1.676

700

75
200

100

1.000
1,776

800

RsiNrORCElfENT

AKD Steel Stbbsses.

Location.

Gauge
Point.

Design T^d
160 lb. per sq. ft.

St. Louis Code
380 lb. per sq.ft.

L. L.

D.L.

Total.

L. L.

D.L.

Total.

Bottom of Slab at
Center

I>ong Span
Short Span

122
113

3,600
7,500

2,000
3.600

5,500 ' 19,500
11.000 , 26,600

2.000
3.500

21.500
30,000

Long Span^

210

3,600

1.600

6.000

37,600

1.600

39.000

Top of SUb at
Support

Short

l-ixea
End

201

6.600

3.600

10,000

28.600

3.600

32.000

Span

Free
End

219

7.600

4.000

11.500

8.600

4.000

12.600

Short Span

101 &
103

7.000

3.600

10.600

17,600

3,600

21.000

Bottom Flange of

Long

Spandrel
Beam

128

3.600

2,000

5,600

9.500

2.000

11,600

I Beam

Span

Interior
Beam

124 &
126

7,600

4,000

11,600

22.000

4.000

26.000

load.

Note. — Stresses due to dead load are obtained by projecting the stress curve to the soro

Digitized by

Google

Report of Committee on Reinforced Concrete. 157

the two sets of ribs can be reached from the data obtained in this
one testy although the stresses obtained indicate that about

nMBSSR^^^^^Sfl^^tSimrS^HIi^

FIQ. 73. — VARIATION OF STRESS AND LOCATION OF POINT OF INFLECTION ALONG
MIDDLE BAR IN DIRECTION OF LONG SPAN.

65 per cent of the load was carried across the short span of the
panel, instead of the 55 per cent assumed.

Digitized by

Googk

DISCUSSION.

Mr. Ash. Mr. L. R. Ash. — I would like to know if experiments have

been made to determine whether the modulus of elasticity of high
carbon steel is really more than that of medium steel. Some claim
that high carbon steel should be used because of the increased
modulus of elasticity.

One other question is whether experiments have been made
to determine the distribution of stresses in a continuous slab with
a concentrated load. A problem that frequently comes to the
bridge designer is how to take care of a continuous slab which has
large dimensions one way and small dimensions the other, for
example a slab supported between two continuous stringers.

Mr. Talbot. Mr. Arthur N. Talbot. — ^The modulus of elasticity of

steel is a fairly definite property. It varies from 29,000,000 lb.
per sq. in. to 31,000,000, perhaps; the average of tests would
bring it a little imder 30,000,000 lb. per sq. in. I do not know
that anything has been foimd which explains or gives the causes
of this slight variation. I am quite sure, however, from tests
which we have made, that there is no appreciable difference
in the modulus of elasticity between the high carbon and the mild
steel.

It is an important question to determine to what extent it
is necessary to reinforce a slab laterally when a concentrated
load is carried. We have been making tests along this line at
the laboratory of the University of Illinois for three years and we
hope to have results to give out soon. I may say now that I
have been surprised to find how little reinforcement is necessary
laterally in distributing the stresses sidewise. In these tests we
have used beams 30 and 48 in. and longer, with depths of 4 to 7
in. and with widths of 30 to 48 in. These slabs were loaded in
several ways, by applying the load entirely across the beam, by
appljring it over half the width of the beam and also over 1/5 of
the width and over 1/10 the width. In a general way it may be
said that with as much as ^ of one per cent of lateral reinforcement
the stresses are distributed laterally over a width of say 8 or 10
times the depth of the beam.

(168)

Digitized by

Google

Report of Committee on Reinforced Concrete. 159

Mr. W. p. ANDEBfiON. — In making the comparison with the Mr. Andenon.
bending moments, what distance was used for the span.

Mr. Talbot. — ^A definite length had to be assumed. It was Mr. xaibot.
thought best to use something more than the clear span and
3 in. more than the clear span for both girders and beams was
used. This is somewhat shorter than that given by the usual
methods of design.

Mr. W. K. Hatt* (By Letter).— The writer has read this Mr.Hmtt.
important contribution with a great deal of interest. .

It is realized that our methods of computing the strength of
a continuous construction of slabs, Boor beams, and columns of a
reinforced concrete building is conventional and very largely
empirical. We need many such tests as are described in this
report to determine the allowable limits of deflection and the real
factors of safety obtaining in the usual designs. The common
practice of loading one panel is evidently misleading since sur-
rounding panels under load assist in carrying the load.

The report confirms the opinion many have had that under
ordinary working loads the tensional stresses in the concrete
assist the reinforcement in carrying the bending moment and that
the conventional computal stress of 16,000 lb. will not be found in
the test floor. Of course imder this distribution the compressional
stresses will be the critical stresses.

The ratios between the compressional stresses and the com-
puted reinforcement stresses in concrete beams reinforced with
various percentages of metal is an important factor in the work
of the designer, who is controlled by building laws. This report
does not comment on the proper values of this ratio. The indica-
tion from the tests, however, is that in spite of high compressional
stresses failiu-es in compression are not evident. The recent pub-
lication of the Bureau of Standards by Messrs. Richard L. Hum-
phrey and Louis Losse, fixes an extreme fiber stress in compression
of 1000 lb. per sq. in. at the measured unit stress of 16,000 lb. per
sq. in. in the reinforcement for 1-2-4 concrete.

The writer made a careful test of the Franks building in
Chicago in the siunmer of 1911 and submits report of this test
as a contribution to the discussion of this subject.

*PYofeflior of Civil Engineeriiig, Pardue University. Lafayette. Ind.

Digitized by

Google

160 Report of Committee on Reinforced Concrete.

REPORT ON TEST FOR ACTUAL STRESSES OF THE
A. J. FRANKS BUILDING, CHICAGO, ILLINOIS.

Bt W. K. Hatt, Consulting Engineer, Professor of Civil Engineering,
Purdue University.

During the latter part of August, 1911, the undersigned loaded panels
of a reinforced concrete building. Fig. 1, constructed by the Leonard Con-
struction Company, for Mr. A. J. Franks, and measured the actual deforma-
tions of the concrete and steel under load. The test was performed upon the
basis of specifications prepared by the undersigned.

In brief, the panels were loaded with pig iron in increments, and the
accompanying deformations were measured in the steel and in the concrete
at all critical points, with a view to fixing a safe limit of loading and to under-
standing the mechanical action of the structure.

TEST STRUCTURE.

The building tested is a ten-story and basement warehouse intended
for the printing and paper trades, Dwight Bros. Paper Company, tenants.
The type of construction used was the Cantilever Flat Slab System, the rein-
forced concrete setting drawings and shop details being made by the Concrete
Steel Products Company, Engineers, of Chicago. The architects of the
building were Richard E. Schmidt, Garden and Martin. The design load
on the floors was taken at 250 lb. per sq. ft. The panel dimensions were
19 ft. 4 in. by 20 ft. 3 in. The four panels under observation were in the
interior of the building and on the tenth floor where the columns were of
minimum size. As the effective clear span between capital and the eccentric
action on the columns were the greatest, the location was such as to insure
the most severe test possible.

OBSERVERS AND METHODS OF OBSERVATION.

The observations were made by experienced observers as follows: Pro-
fessor H. H. Scofield, of Purdue University; Professor W. A. Slater, of the
University of Illinois Engineering Experiment Station; Mr. W. E. Ensign,
of the University of Illinois Engineering Experiment Station; with the assist-
ance of Professor L. W. Weeks, of Purdue University.

The deformations were determined by the use of extensometers of the
type devised by Professor H. C. Berry of the University of Pennsylvania.
On the steel a gauge length of 10 in. was used, which, with the multiplying
lever in the instrument, gave direct readings of unit deformations of .00002
in. per in., corresponding to stresses of 600 lb. per sq. in. in the steel, and
it was possible to estimate clearly fractions of this amount. On the concrete
readings the gauge length was 6 in. and the direct reading of unit deforma-
tions was .000033, corresponding to a stress of 133 lb. per sq. in., and it was
possible to estimate fractions of this amount with accuracy.

Errors in operating the instrument were reduced to a minimum by tak-
ing every reading at least five times and by calibrating on a standard bar at

Digitized by

Google

Report of Committee on Reinforced Concrete. 161

frequent intervals. Four standard bars were used throughout the test, all
of these being embedded in the concrete and subject to the same temperature
changes as the rods in the test floor but free from any stress due to applied
load. By reading on these bars between sets of five or six test readings the

FIG. 1. — A. J. FRANKS BUILDINQ IN COURSE OF CONSTRUCTION,
CHICAGO, ILL.

observations on the materials under test were freed from temperature differ-
ences and systematic errors.

The deflections were measured to .0001 in. by use of the deflectometer
described in a previous test fQr actual stresses.

Digitized by

Google

162 Repobt of Committee on Reinforced Concrete.

Tensile defonnations were measured throughout the test over 42 gauge
lengths on the steel reinforcing rods; compressive deformations in the con-
crete were measured over 26 gauge lengths; deflections were observed at 24
points, and 27 other readings of deformations were taken throughout the test
to study such phenomena as the arch and slab action, the distribution of

FIQ. 2. — ^LOCATION OF GAUGE LBNGTHS IN CONCRBTB BLAB.

stress, and the eccentric loading on the columns at the edges of the loaded
area. The location of the gauge lengths is given in Figs. 2 and 3.

The observations were arranged in groups, each designed to cover
adequately some particular feature, and synmietrically located observations
were obtained as a check in every case where possible, in order to cover varia-
tions in the quality of the concrete at different points.

Digitized by

Google

Report of CoMBflTTEE on Reinforced Concrete. 163

The conditions were exceptionally favorable to a satisfactory test.
The building was practically completed at the time of test and the variation
in temperature was very slight, being about 74** at the start of the test,
dropping gradually and uniformly to 70°, and rising again at the end to 74°.

FIG. 3. — LOCATION OF GAUGE LENGTHS IN REINFORCING RODS.

The slab was poured on June 23, 1911, and was 64 days old when the maximum
load of 624 lb. per sq. ft. of panel area was placed upon it.

The amount of loading was determined by the weight of the pig iron as
recorded on the weigh bills delivered by the teamsters, and was checked by
weighing a number of piles of pig iron from the test load on a platform scales.

The pig iron was piled on the floor in separate piers, Fig. 4, each placed

Digitized by

Google

Digitized by

Google

Report of Committee on Reinforced Concrete. 165

within a two-foot square, so that no arch action existed in the load itself to
relieve the panel from moment. It was necessary to leave several of these
squares vacant about columns and to leave an aisle between the columns to
allow space for observations. The pig iron belonging on these vacant squares
was distributed over the remaining squares of the same panel. It will be
apparent that under this procedure the real load intensity, as affecting bend-
ing stresses and deflections, is the intensity of loading over the loaded area
(about 90 per cent of panel area), rather than the nominal or average load
over the entire panel area. It is evident that had the 10 per cent of panel
area close to the columns been loaded to the same intensity the increase in
stresses and deflection would have been practically nil.

All gauge lengths were measured and checked throughout before loading
was started. When the loading reached certain amounts, distributed evenly
over four panels, it was discontinued and allowed to stand for 6 hours before
the observations were made and the loading resumed. The increments of
load at which measurements were made were as follows: 75, 150, 256 (the
design load), 312 (intensity, 359), and 624 (intensity, 717) lb. per sq. ft.,
the latter load being applied to two panels only. Readings were also taken
with 256 on two diagonal panels and 312 on the other two panels, and with
468 on two diagonal panels and 156 on the others, but without waiting for
the six-hour interval to elapse before taking readings. The total number
of complete observations over single gauge lengths was over 2,000, and the
total individual readings over 10,000.

STRESSES.

The stresses were determined from the observed deformations by using
a modulus of elasticity of 30,000,000 lb. per sq. in. for the steel, and 4,000,000
lb. per sq. in. for the concrete. The latter value was determined from tests
of three concrete prisms poured from the concrete in the test slab and tested
at Purdue University at an age of 77 days.

Table I gives a summary of the corrected values of the total dead and
live load stresses observed in the various groups of observations. The detailed
summary of individual stresses at the various observation points are omitted
in this report. After the observations had been corrected for temperature
and observational errors, by use of the standard bar calibrations, load deforma-
tion curves were plotted for each observation point, the known nature of the
load deformation curve under flexure being used as a basis. The dead load
stress has been taken from these curves as equal to the stress caused by an
equal live load.

COMMENTS ON THE RESULTS.

The undersigned is not prepared at the present time to state the signifi-
cance of the results obtained with respect to the mechanics of this form of
construction. Such statements would have somewhat of a speculative ele-
ment and should be separated from the report of test which is one of measured
facts.

Digitized by

Google

166 Report of Committeb on Rbinfobced Concrete.

As compared with the design requirements of the Chicago Building
Code, it is interesting to note that at design load the highest average stress
in the steel is 5078 lb. per sq. in., while the highest average compressive

.ft.

Table I. — Condensed Summary of Actual Stresses.

All stresses are the final values for total dead and live load, and are given
in lb. per sq. in.

Stresses in Slab Rods.

oJt^JSS.tc., DeBcription. ^3^

5, 6, 7, 14, 15, 16 Center of span, diagonal band 1071

8, 9, 10, 11, 12, 13 Center of span, cross band . . 4539

31,32,33,34,35,37 Over capital at center col-
umn, diagonal band 3440

27, 28, 29, 30 Over capital at center col-
umn, cross band 4575

21,22,23 Over capital at comer col-
umn, diagonal band 1920

24, 25, 26, 41, 42, 43 Over capital at side column,

cross band 2690

Live Load-lb. per 0q. f
812 624

1920

6140

10095

4350 9280

4840 8140

2280 7540

3138 5315

Compressive Stresses in Concrete.
66, 67, 69, 60, 68, 69, 70. .On slab at center column. . .

61, 62, 65, 66 On drop at center column. . .

61, 74 On drop at comer column . . .

52, 53, 73 On slab at corner column ....

54, 55, 71, 72 On slab at side columns

560 650 1206

677 778 1685

318 370 1515

329 378 650

189 217 420

Mcudmum Stresses in Columns Due to Eccentric Live Load.

104 . . Compression in concrete, corner column 680 840 1660

109. .Compression in concrete, side column 416 512 1000

84. .Tension in steel, comer column 4980 6000 11620

99. .Tension in steel, side column 2220 2640 5880

Deflections in Inches.

121, 124. .Center of panel— at 6 hours 123

Center of panel — at 24 hours

After standing unloaded 6 hours 142

After standing unloaded 2) days 090

.156

.475
.500

stress in the concrete is 677 lb. per sq. in. On the basis of safe working
stress in the steel of 16,000 lb. per sq. in., and in the concrete of 35 per
cent of the ultimate strength (which averaged over 3250 by tests of prisms)
or 1100 lb. per sq. in.; it appears that the steel is stressed to 31 per cent

Digitized by

Google

Report op Committee on Reinforced Concrete. 167

of its safe load while the concrete is stressed to 62 per cent of its safe load.
It appears, therefore, that the design is overbalanced with an excess of steel.
At the highest applied load of 717 lb. per sq. ft., the ratio of the steel and
concrete stresses remains practically unchanged.

The eccentric action of the test load was most marked on the comer
columns of the loaded area and was suj£cient to produce a tension in the
steel of 5000 lb. per sq. in.

With respect to the strength of the structure, it may be said that a nominal
load of 624 lb. per sq. ft. of panel, actually 717 lb. per sq. ft. of loaded surface,
was applied without producing any permanent damage to the building. At
this load the highest observed average total dead and live load stresses were
less than 12,000 lb. per sq. in. in the steel and less than 1700 lb. per sq. in.
on the concrete.

From a consideration of the data from the above test the writer con-
cludes that the A. J. Franks Building is amply strong to carry the designed
load and that the lower floors at least may safely and continuously be loaded
with considerably more than the designed load.

(Signed) W. K. Hatt.

Digitized by

Google

THE TESTING OF REINFORCED CONCRETE
BUILDINGS UNDER LOAD.

By W. A. Slater.*

I. Introduction.

Development of Building Tests. — ^For several years there has
been a growing demand for tests of full-size structural members.
A more recent development is the test of structures themselves
and the measurement of actual stresses in the component parts.

Load tests have been required by city building departments
as a condition of acceptance of reinforced concrete buildings and
have been used by construction companies and engineers to
demonstrate the adequacy of various designs. Such load tests
are never continued to destruction, the applied load being gen-
erally twice the design live load, and emphasis is placed upon
measurement of deflection and recovery. No measurements of
stresses are made in such tests and under these conditions the
safe load can not be fixed upon as a definite ratio of the ultimate
load. The deflections observed in such tests constitute a very
inadequate and actually misleading measure of the stresses.
Slight deflections have been taken to indicate low stresses in
reinforcement and in concrete, but recent tests in which deforma-
tions were measured have shown that even with slight deflections
large stresses are developed in concrete even when the reinforce-
ment stresses are low. The tendency of building codes was to
disregard continuity of action in beams in reinforced concrete
buildings and to specify the design as of simple beams, but even
in such cases a small amount of reinforcement was placed across
the support to prevent the opening of large cracks. This rein-
forcement and the tensile strength of the concrete have been
sufficient to develop a large stress in the concrete at the support
which may not have been specifically provided for. Thus the
so-called conservative attitude of not allowing anything for
continuity of beams at the support may prove a source of weak-

* Engineering Experiment Station, University of Illinois. Urbana. 111.

(168)

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 169

The measurements of deformations in building structures
eonfirms the truth of this statement.

Purpose and Scope of this Paper. — ^The reports of all building
tests in which deformations have been measured deal in the main
with the behavior of the structure and record the results, and are
not primarily concerned with the working of the instruments or
with the methods of making the tests. To conduct a successful
building test is diflScult, however, and this paper is written in
order to present information as to methods of testing gained by
experience and to point out certain respects in which such tests

Building

euf'Mhff

CorMfon Bolklfng

•RELATIVE SIZE OF FLOOR AREAS TESTED.

may be conducted more satisfactorily than those which have
already been made. The following general order of presenting
the material in hand will be observed: (1) enumeration of tests,
(2) the planning and preparation for a test, (3) the instruments;
their construction and use, and the methods of making observa-
tions, (4) the methods of making calculations and (5) the cost of
a test.

The following is a list of tests of building floors in which the
methods described herein of measuring deformation were used.
Fig. 1 shows the range in size of these test areas.

Digitized by

Google

170 Slater on Testing Reinforced Concrete Buildings.

Test No. 1. — Deere and Webber Building, Minneapolis,
Minnesota, October and November, 1910; flat slab floor with
four-way reinforcement; built by Leonard Construction Com-
pany of Chicago, and tested by them with the co-operation of
the Erigineering Experiment Station of the University of Illinois.

Test No. 2. — ^Wenalden Building, Chicago, Illinois, June and
July, 1911. Beam and girder building constructed by Ferro-
concrete Construction Company of Cincinnati, and tests made
by co-operation between the National Association of Cement
Users, the Ferro-Concrete Construction Company and the Engi-
neering Experiment Station of the University of Illinois.

Test No. 3. — ^The Powers Building, Minneapolis, Minnesota,
July and August, 1911; flat slab floor with two-way reinforcement;
built and tested by Corrugated Bar Company of St. Louis.

Test No. 4. — ^Franks Building, Chicago, Illinois, August,
1911; flat slab floor with four-way reinforcement; buUt and
tested by Leonard Construction Company of Chicago.

Test No. 5. — ^Turner-Carter Building, Brooklyn, New York,
September, 1911; beam and girder floor; built by Turner Con-
struction Company of New York; test made by co-operation
between National Association of Cement Users, the Turner
Construction Company and the Engineering Experiment Station
of the University of Illinois.

Test No. 6. — Carleton Building, St. Louis, Missouri, October,
1911; flat slab floor with two-way reinforcement; built and
tested by Corrugated Bar Company.

Test No. 7. — ^Barr Building, St. Louis, Missouri, December,
1911; full size test panel (25 ft. x 26 ft. 9 in.). Terra-cotta tile
used to lighten construction; gives two-way T-beams with web
between tile on tension side and concrete flange above the tile;
two-way reinforcement. Panel built by Corrugated Bar Com-
pany to demonstrate eflSciency of design proposed for Barr Build-
ing in St. Louis; test made by Corrugated Bar Company.

Test No. 8. — Ford Motor Building, Detroit, Michigan,
February and March, 1912; flat slab floor; built and tested by
the Corrugated Bar Company.

These seem to be the only full-size reinforced concrete floor
tests on record in which deformations in reinforcement and con-
crete have been measured. The writer was in immediate charge

Digitized by

Google

Slater on Testing Reinforced Concrhtb Buildings. 171

of Tests Nos. 2 and 5 and had a part in tlie conduct of 'all "tjie
others except Tests Nos. 6 and 8. I'jie 'motfioas' of t&3tiag
presented in this paper were developed by the writer as a , result
of his connection with the tests. These meth'ods*v>^re ji^dft^^'V^
increase the accuracy of results, to avoid accidental^ errors^ ai».d t^
correct for systematic errors.

Much credit for the initiative in this type of test ife due Mr.
A. R. Lord, formerly research fellow at the University of Illinois,
who was largely instrumental in bringing about the test of the
Deere and Webber Building, the first in the series named. After
the presentation of Mr. Lord's paper on the test of the Deere and
Webber Building, The National Association of Cement Users
decided to continue the investigation. All of the tests given in the
above list were conducted on the same general lines as that of the
Deere and Webber Building. Only the tests of the Wenalden
Building and the Turner-Carter Building were in the series author-
ized by the National Association of Cement Users, but the results
of the tests made by the Corrugated Bar Company on the Powers
Building and on the Barr Building test panel have been placed
at the disposal of the Association. The Franks Building test,
made by the Leonard Construction Company, was an investigation
planned to give data for an intelligent modification of the Chicago
Building Code. The other two tests, those of the Carleton Build-
ing and the Ford Motor Building, were in the nature of investiga-
tion of special features of design. The methods used in all of
these tests are essentially the same and have been developed
at the University of Illinois Engineering Experiment Station.

Available Literature, — Reports of results of some of these
tests are available as follows:

1. Deere and Webber Building.

Paper by A. R. Lord, "A Test of a Flat Slab Floor in a
Reinforced Concrete Building." ProceedingSf Vol. VII,
1911.

Abstracts: Engineering News, December 22, 1910; Engi-
neering-Contracting, December 22, 1910.

2. Wenalden Building and Turner-Carter Building.
Report of Committee on Reinforced Concrete and Build-
ing Laws, Part I. Proceedings, Vol. VIII, 1912.*

♦See pp. 61-167.— Ed.

Digitized by

Google

172 Slater on Testing Reinforced Concrete Buildings.

• • • • •

• •• •'

Abilrkctfi f \Fngin6efi7iy. i?6cord, March 23 and April 13,
• le/^y i^inf 4r«W Vf ^w's, April 18, 1912; and CetnerU

••\i: : .. . :-4ffV^ei;w *•*

•/ ; S?*! JVveiiHuiiaihg and Barr Building Test Panel.
. . .'. *. *.llep6y; g/ tlgmi^ttee on Reinforced Concrete and Building
;: V: ( } V Law)a;-i^arteil and III. Proceedings, Vol. VIII, 1912.*
Abstracts: Engineering News, April 18, 1912; and Engi-
neering Record, April 20, 1912.
4. Franks Building.

(a) Abstracts of paper by W. K. Hattf before Indiana
Engineering Society, Engineering-Contracting, March
13, 1912; and Engineering News, April 8, 1912.
(6) Discussion by W. K. Hatt on Report of Committee on
Reinforced Concrete and Building Laws, Proceed-
ings, Vol. VIII, 1912.t
(c) Trade publication on Cantilever Slabs published by
Concrete Steel Products Company, Chicago, 111.

II. Conduct op Tests.

DEFINITIONS.

In the following descriptions of tests, many terms will be
used for which somewhat arbitrary definitions will need to be made.
These definitions are:

Gauge Hole: A small hole (.055 in. is here recommended)
drilled into the steel bar or into the plug inserted in the concrete
has been termed a gauge hole. It is for the admission of the point
of a leg of the extensometer.

Gauge Line: The gauged length connecting a pair of gauge
holes is termed a gauge line.

Reading: A reading is a single observation on any gauge line.

Observation: An observation as here used is the average of a
number of readings.

Zero Length of Instrument: The length of the instrument at
the time of taking the first observation on the standard bar will
be known as the zero length of tlie instrument. This first observa-

• See pp. 61-157.— Ed.
t See pp. 159-167.— Ed.

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 173

tion on the standard bar is not the zero length, but a comparison
of a subsequent observation with it shows any change from the
zero length.

Correction: A correction is the amount which if added alge-
braically to the observation will give the observation which would
have been obtained if the instrument had not changed from its
zero length.

Series of Observaiiona: The observations taken consecutively
at a given load without repetitions on any gauge line ar^ defined as
a series of observations.

Interval: An interval as used here is the time elapsing between
consecutive observations, and all intervals in *any series are (for
lack of more exact information) assumed to be equal. For this
piupose the average of two consecutive observations on standard
gauge lines is considered a single observation.

Standard Gauge Line: This is a gauge line used usually to
determine changes of length of instrument, of reinforcement or of
concrete due to other causes than the applied load. Its purpose
usually is to determine the temperature effect on the instrument,
but it may be used to detect accidental changes of instrument or
temperature stresses in the reinforcement of the concrete. Origi-
nally this gauge line was placed on a steel bar separate from the
structure, and this gave rise to the term standard bar. In several
of the later tests, however, the standards have consisted of gauge
lines placed in the reinforcement and concrete of the structure
remote from the area affected by the load. Standard gauge line
is adopted, therefore, as the more general term and any reference
to the standard bar may be understood to signify the standard
gauge line on a bar separate from the structure.

general outline of method of testing.

After determining what measurements will best give the
information desired from the test, the gauge lines are laid off on'
the surface of the concrete and small holes are cut or drilled in the
concrete at a predetermined distance apart in order to expose the
reinforcement or allow a metal plug to be inserted, according as
the measurement is of reinforcement or concrete deformation.
The metal plugs used are securely held in place by imbedment'

Digitized by

Google

174 Slater on Testing Reinforced Concrete Buildings.

in plaster of Paris. The gauge holes having been carefully pre-
pared; a set of zero readings is taken on all gauge lines, an increment
of the loading material is then added and a second series of observa-
tions on the gauge lines taken. The difference between the two
readings on the same gauge line represents the deformation in
that gauge line. It is possible that this apparent deformation
may be due partly to temperature changes in the instrument
instead of stress changes of the material by reason of applied load.
For this reason reference measurements are made on standard
unstressed bars made of Invar steel which has a very low coefficient
of expansion and whose change in length due to temperature
would therefore bfe very sUght. From these readings on the
standard bar, temperature corrections are computed as shown in a
later paragraph and applied to the observations in order to deter-
mine the actual change in length of the gauge line. Another
increment of load is then applied and another series of observations
taken.

Floor deflections also have been measured in all of these tests,
but they have been considered as of secondary importance. They
have been used to throw light on the correctness or incorrectness
of the deformation readings and to gain some idea of the general
distribution of stresses throughout a floor. They can apparently
be depended upon to show with considerable accuracy the pro-
portional rate of increase of stress, but deflection formulas are so
imperfect that measurement of deflections can not be depended
upon to give the actual values of stresses.

Measurements of dimensions such as span, depth of beams,
location of observation points, weight of loading material, location
of cracks, and any other measurements which were considered of
value in working up results have been carefully taken. The
measurements taken are usually distributed over and under the
surface of the floor tested in order to gain an idea of the changes
occiuring in different parts of the structure.

The above statement gives in general terms the features of
any one of the tests dealt with in this paper. There are many
difficulties to overcome and many chances for error. What
follows is concerned mainly with the method of overcoming these
difficulties and avoiding these errors. Most of the statements
made represent the results of experience on previous building

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 175

tests. Some merely give ideas which it is believed if put into
operation would be advantageous.

the planning op a test.

Each test made will involve individual consideration of the
choice of area to be loaded, the number and location of gauge
lines and deflection points, the number of laborers required, the
loading material to be used and its distribution, and provisions
for storage of the loading material near the test area without
appreciably affecting the stresses which are to be measured.
Other matters will come up for consideration, but in the main
they will not require different solutions for each test.

Choice of Test Area. — The area to be loaded should be chosen
so as to fulfil the following conditions as completely as possible:

(a) It should be so located as to give conditions in the beams,
slabs, columns, etc., as severe as will be found anywhere in the
building when in use.

(6) It should be free from irregularities of construction.

(d) It should be as easily accessible to the loading material
as possible.

In most cases some limitation is found on part or all of the
conditions named. For example, in the test of the Wenalden
Building it was impossible to find an area entirely free from irregu-
larities of construction. An industrial track crossed one of the
panels chosen, and the floor was thicker immediately under this
track than at other places. On the edge of one or two of the panels
tested, beams about an inch deeper than the regular beams were
located. However, none of the measurements assumed to give
typical results were taken in these panels, and it is believed that
the stresses in the other panels were not affected appreciably by
these irregularities. Again, in the test of the Franks Building
it was not possible to choose a lower floor convenient to the loading
material. An upper floor was used in order, during the course of
construction, to make preparation for the test, thus avoiding dig-
ging in the concrete. However, this choice of a floor fulfilled one
of the conditions mentioned, in that it gave a much more severe
test of the columns than a test on a lower floor would have done.

Digitized by

Google

176 Slater on Testing Reinforced Concrete Buildinqs.

Also, in the test of the Carleton Building at St. Louis the area to
be tested was specified by the city building department, and there
was no choice as to location on the part of those making the test.
Number oj Measurements. — ^The number of measurements to
be taken will depend upon the nature of the test, the number of
observers, and the number of laborers. If the test is a part of a
series by which it is expected to gain scientific information which
will afford a basis for design, it is likely that it will be deliberate
enough that a large number of measurements may be taken.
Such tests were those of the Wenalden Building, the Franks
Building, the Turner-Carter Building, and the Barr »test panel.

Pl/^n5how/a4 5 Location ^ 1 > f>L/9AfSMO¥Y*MgLoCATtON

o^ fio/Afrs OAf Top OF Slab "^y^ or Po/Afrs cr^Sorromor Sl^B

•• M[>inf on sfevi '

0aQg0l9nff/i 8 tnc/»99 I

FIG. 2. — ^LOCATION OF GAUGE POINTS, CARLETON BUILDING.

If, on the other hand, the test has more of a commercial
nature or is a utilization of the opportunity offered by the accept-
ance test to take some measurements which will show actual
stresses, or if for any other reason the test is hurried, the number
of measurements will necessarily be rather small. Of this class,
the tests of the Carleton Building in St. Louis and of the Ford
Motor Building in Detroit, Michigan, are good examples. Notice
was given the engineers only about one day in advance that a
test would be made on the Carleton Building. Permission was
obtained from the contractor to expose bars for measurement in
various points and to erect the necessary scaffolding. The meas-

Digitized by

Google

Slater on Testing Reinforced Concrete BxnLDiNGs. 177

urements were made more for the purpose of checking the anal-
ysis upon which the design was based than to form in itself a
basis of design. Therefore comparatively few observation points
were used. It is believed that this test is representative of the
type of test which is practicable on a commercial basis, hence (by
courtesy of the Corrugated Bar Company) a plan is given in
Fig. 2 showing the points where measurements were taken.

Disirihulwn of Measurements. — The arrangement of observa-
tion points will depend on what are the principal subjects for
investigation in the test. Whatever the subject of study may be,
the observation points should be arranged in such a way that a
curve of deformations may be plotted against distance, showing

I ^ 5 S §
§ § § § §

§ ^ S ^ §

^ S S

Deformat/on per Unit of Length.

PIG. 3. — LOAD-DEFORMATION DIAGRAM FOR SERIES OP GAUGE LINES ON
REINFORCEMENT, POWERS* BUILDING.

a gradual progression from the condition at one part of the struc-
ture to the condition at another, for it is found that there are even
imder the most careful work, inconsistencies which will make the
results look doubtful if standing by themselves. The points so
arranged should be numerous near the place where the measure-
ments of greatest importance are to be taken, so that the results
will not depend upon measurements at a single point, or upon the
average at portions of the structure supposed to be similarly
situated but in different parts of the building where unknown
conditions may actually cause a large variation in the phenomena
of the test. It will not be possible to carry out this plan for all
subjects of investigation, as the number of observations required

Digitized by

Google

178 Slater on Testing Reinforced Concrete Buildings.

would usually be impracticably large. Such provisions may be
made to cover the main lines of investigation, and isolated obser-
vation points may be used to gain information as to tendencies
of other portions of the structure, but of course, less reliance must
be placed on the results of the latter measurements than where
the larger number of observations is made. It would be advan-
tageous, as was done in the Powers Building test and also in the
Barr test panel, for two observers to check measurements on the

FIG. 4. — LOCATION OF GAUGE LINES, POWERS BUILDING.

same points. One or both of these checks is very valuable in
establishing the correctness of observations.

Figs. 3, 4, 5, and 6 illustrate the former method. Fig. 3 gives
the load deformation diagrams for several gauge lines in the test
of the Powers Building. Fig. 4 shows the location of these points
with reference to the wall and a column. Fig. 5 shows the same
data plotted as deformation against distance from the column
instead of against load. It may be seen that the correctness of
the load deformation curve for one of these points, if standing by

Digitized by

Google

Slateb on Testing Reinforced Concrete Buildings. 179

itself, might be doubted because of the complete change in the
character of the curve at a load of 200 lb. per sq. ft. But when
these deformations are plotted against distance, the results look
so consistent that it is scarcely conceivable that they are seriously
incorrect. In the test of the Wenalden Building very high
deformations were observed in the concrete of the beams near the
supports; so high that the results were doubted, and as the points
on the load deformation curves were few and scattering, there
was often room for doubt. For this reason it was considered
especially important that evidence which would confirm or dis-
prove this high compression in the concrete be obtained in the

V .oooa
g ^.oood

^.0003

^.oooz

.0001

. D /stances from Capital in Inches.

PIG. 5. — DATA OF FIG. 3 PLOTTED AS A DISTANCE-DEFORMATION DIAGRAM,

POWERS BUILDING.

test of the Turner-Carter Building; accordingly the method of
placing observation points at frequent and regular intervals along
the ends of the beams was used. The deformations measured
are plotted in Fig. 6 against the distance from the supporting
column, and the results not only tend to show the correctness of
these measurements but also to indicate that the high stresses
observed in the beams of the Wenalden Building were actually
present.

Subjects of Investigation, — In the tests discussed in this paper
deformations have been measured with a view to obtaining
information on each of the following subjects:

Digitized by

Google

180 Slater on Testing Reinforced Concrete Buildings.

(a) The values of the moment coefficients at the center and
support of the beam or slab under mvestigation.

(6) Relative moments at support for various conditions of
fixity.

(d) Bond stresses.
{e) Diagonal tension.
(/) Stresses in columns.

(gf) Time effect under constant load.

FIG. 6. — DISTRIBUTION OF COMPRESSIVE DEFORMATION IN BOTTOM OF COLUMN
BEAM, TURNER-CARTER BUILDING.

(A) The lateral distribution of stress to parts of the structure
entirely outside of the loaded area.

(i) The extent to which reinforcement stresses are modified
by errors in the assumption that no tension is carried by concrete.

(j) Stresses in slabs of beam and girder construction.

Other subjects of investigation have received attention, but
these are the most important ones. Some phenomena have been
observed, offering additional problems, of which the determina-
tion of the amount of arch action present is probably the most
important. It is important both in itself and because it is

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 181

intimately involved in the determination of moment coefficients.
A further discussion of this problem is here given.

The attempts to determine moment coefficients have not
been entirely successful due to errors of measurement and unex-
pected variations in similar parts of the structure remote from each
other. The method has been to measure deformations on both
reinforcement and concrete at the center and support, and from
these measurements to determine the total resisting moment
developed. Equating this resisting moment to a constant K
times Wl a solution is made for the value of K, The indications
that arch action has been present have so complicated this that
even where measurements have appeared quite satisfactory, the
uncertain amoimt of arch action entering has rendered the value
of K uncertain.

A proposed method of determining the amount of arch action
in any case is to make a special study of the deformations in a
cross-section at the center of each beam across an entire panel.
In this study, deformations should be observed on the reinforce-
ment and at various elevations on the concrete so that the position
of the neutral axis and of the center of gravity of tensile and com-
pressive stresses respectively can be accurately located without
dependence upon the law of conservation of plane sections. By
this means it should 'be possible to determine if the sum of the
compressive stresses is in excess of that of the tensile stresses.
If so, the difiference apparently must be the direct thrust due to
arch action. The same study can be made, though not so satis-
factorily, at the ends of the beams. This measurement of thrust
will require observations on an extremely large number of gauge
lines, and it would appear important to concentrate the greater
part of the attention of the test on one panel.

If the floor be considered to be made up of strip-beams of
differential width capable of transmitting shear from strip to strip,
it is not necessary, for perfect beam action, that the sum of the
tensile and compressive stresses on a cross-section of any one strip
be zero. However, beam action does require that the sum of the
tensile and compressive stresses on the total cross-section of the
beam should be zero, and for this reason it is important to extend
the investigation sufficiently to determine if appreciable deforma-
tions are continued out into the panel adjacent to the loaded area.

Digitized by

Google

182 Slater on Testing Reinforced Concrete Buildings.

Laborers, — The number of laborers which can be used advan-
tageously will depend on the distance from which the loading
material is to be transferred, on the size and accessibility of the
tested area, on the amount of work which can be done by them
during the intervals between increments of loading while observa-
tions are being taken, and on the length of time required to take
a series of observations. The handling of the labor should, if
possible, not be left to the one in charge of the test, as proper

FIG. 7. — BRICK AND CEMENT AS A LOAD, WENALDEN BUILDING.

attention to the conduct of the test demands all of his time. In
the tests included in this paper the number of laborers has varied
between wide limits, from 5 or 6 in the Powers test to 30 or 35
in the Deere and Webber test.

Loading. — ^In the tests which have already been made, the
following loading materials have been used: brick, cement in
bags, loose sand in small boxes, sand in sacks and pig iron. The
material used will almost always be that which is most easily
available, because the moving of loading material from any

Digitized by

Google

Slater on Testino Reinforced Concrete Buildinqs. 183

distance adds very greatly to the cost of the test. Leaving con-
sideration of cost out of the question, sand in sacks seems to be the
most satisfactory of the materials above mentioned for loading
purposes. Some of the qualities of the materials mentioned are
as follows:

(a) Brick: Brick spalls and chips in handling, covering the
floor with dust and jagged particles which cause discomfort to the
observer in kneeling to take observations. It is important to

PIG. 8. — SAND IN BOXES AS A LOAD, TUBNER-CABTER BUILDING.

avoid this because discomfort necessarily decreases the accuracy
of his observations. This might be avoided by sweeping, but in
sweeping it is diflScult to avoid getting dirt into holes where obser-
vations are to be taken, and this is just as troublesome as having
the dirt on the floor. Fig. 7 shows the use of both brick and cement
in the same test. Attention is called to the proximity of the cement
sacks to the beams and girders of the floor above. In some cases
the intensity of the load would be limited by the height of the
ceiling if cement and brick are used.

Digitized by

Google

184 Slateb on Testing Reinforced Concrete Buildings.

(6) Cement: Cement sifts through the sacks and the sacks
become imtied, scattering cement on the floor, filling observation
holes and causing much dust in sweeping or cleaning up. The
dust is injurious to delicate instruments and annoying to observers
and recorders.

(c) Loose Sand in Small Boxes: As sand is usually damp, it
does not have the fault of causing dust and consequently is more
easily cleaned up than the other materials mentioned. There are,
however, other objections to it. In filling boxes it is difficult to
avoid spilling the sand around and between the boxes, and con-

FIG. 9. — BAND IN SACKS AS A LOAD, BARR BUILDINQ TEST^PANEL.

sequently filling the observation holes. On account of the great
difficulty in removing loose sand without spilling a great deal of
it, it is impracticable to take observations as the load is being
removed, therefore it is necessary to remove in one increment the
whole load from a given panel. Fig. 8 shows this method of loading,
(d) Sand in Sacks: Sand in sacks constitutes a very satis-
factory loading material. Fig. 9. It was piled up to a height of
about 9 or 10 ft. and very little inconvenience was caused by the
sacks becoming untied or by spillmg the sand. The worst difficulty
encountered, and this exists with all materials handled in sacks,
is that of the slidmg of sacks on themselves when the load is piled

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 185

high. It can be seen in Fig. 9, above referred to, that bracing
was necessary to prevent the sand from sHding together and
filling up the aisles. It is a source of danger to those taking
observations as, if a slide should occur, it would probably give
very little warning and might catch the observer while in such a
position that he could not escape. However, this objection
would be likely to occur with any material which is piled as high
as was that in this test. Under any circumstances it is necessary
that care be taken and undue risks avoided.

i FIG. 10. — pig iron as a LOAD FRANKS BUILDING,

the making of the test, the worst objection to it is that, as with
the brick, small particles break off and cause annoyance to observ-
ers. This is less noticeable than with brick and in other ways
pig iron is clean. It possesses the great advantage that with its
use a very heavy load can be applied without piling the load
extremely high.

Tin plate in boxes 2 ft. square, each weighing 200 lb., was to
have been used in a building test. A more nearly ideal loading
material would probably be hard to find, but unfortunately this
test could not be carried out.

The intensity of the loading will depend mainly on the load

Digitized by

Google

186 Slater on Testing Reinforced Concrete Buildings. ,

for which the building was designed. It will not be possible to
make the load absolutely uniform, as aisles will be necessary for
the purposes of (a) convenience in placing the load, (6) access to
gauge lines for the taking of observations and (c) prevention of

mmk

FIG. 11. — MOMENT AND SHEAR DIAGRAMS FOR
THREE ARRANGEMENTS OF LOAD.

arching in the loading materials. It has been found that it is
difficult to cover more than about 75 per cent of the actual area
of the floor, and in many cases less than this will actually be
covered. Hence in computing the probable height of the load,
this fact must be taken into consideration.

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 187

Aisles should be so placed that the load, even though partly
carried by archmg of the material, will cause stresses in the floor
which are approximately equal to and always as severe as those
caused by an actual uniform load. Fig. 11 shows the moment
and shear diagrams which would be obtained by loading a simple
beam with a total load W distributed over the span in three dif-
ferent ways, as follows:

(a) Solid Line: Uniform load W, over full span.

(6) Broken Line: Same load W distributed over one-half
of span, giving aisles of equal width at center and support.

(c) Heavy Dotted Line: Same load W distributed over
one-half span, half of load being carried by arch action to ends
of boxes (shown here as concentrated loads W/8), and the other
half being uniformly distributed over the half span.

It will be possible in almost any test to arrange the loading
material in such a way as to come within the limits outlined by the
three arrangements of load assumed in Fig. 11, and it is seen that
if this is done, the presence of the aisles or of arching to the sides
of the boxes or piers, while not affecting the amoimt of the maxi-
mmn moment and the maximum shear, would tend to cause them
to exist over greater portions of the span than would be the case
with an equal imiform load. In this figure aisles equal to one-
quarter of the span have been assumed. In no case would they
be as large as this, and, therefore, the moment and shear diagrams
should actually conform even more nearly to those for uniform
load than is shown in the figure.

Arrangement should be made, if possible, to store the loading
material near the test area to hasten the work of applying the
load after the test begins. The general rule has been to allow
loadmg material to be stored as close as one full panel length from
the test area, but the intensity of the storage load has been kept
down as much as possible.

preparation for the test.

Cutting Holes in Concrete. — ^In all of these tests it is necessary
to cut holes in the concrete in order to expose the reinforcement.
Fig. 4 shows a hole cut in the concrete of the Powers Building
where a series of measurements was taken on a rod passing through
a column. This cutting has been best accomplished by the use

Digitized by

Google

188 Slater on Testing Reinforced Concrete Buildings.

of a cold chisel with a very gradually tapering point. This is a task
for common laborers and a long one for inexperienced men, but it
has been found that a great deal of speed can be developed by
practice, hence the importance of completing this part of the work
with a single set of workmen.

A saving in mutilation of floors can often be effected by
planning the test ahead of time and inserting plugs in the concrete
during construction in the proper position for the gauge lines.
Removal of the plugs after the concrete has set exposes the rein-
forcement without the use of a cold chisel. Likewise metal plugs
may be set in the concrete at the proper positions for the measure-
ment of concrete stresses and thus save cutting into the concrete
to place compression plugs. The point has been raised that by
preparation of this kind a chance is given to the contractor to
know what panels are to be tested and thus to make the construc-
tion of that panel better than others. For this reason there is
room for question as to the advisability of using this method. In
most of the tests under consideration this point has been taken
care of by the fact that it was not known until shortly before the
test what area was to be loaded. It is believed that the saving
thus effected is not generally sufficient to justify prejudicing the
test by the use of this method.

Drilling of the gauge holes will be discussed under the subject
''Instruments and Observations. ''

Scaffolding, — A platform supported on some kind of scaffold
is necessary which will enable the observer to get close enough to
the floor above to take observations of deflection and deformation.
This should be at such a height that when the observer stands
upon it the points where measurements of deformation are to be
taken will be about one inch above his head. For flat slab con-
struction this condition is easily obtained (see Fig. 12), but with
beam and girder construction where there are measurements
on beams, girders, and the floor slab, the heights df different gauge
lines are so different that arrangement will need to be made for
building certain parts of the platform higher than others (see Fig.
13). It is important that the elevation of the platform should
be such that the observer can stand erect while taking the readings,
and yet such that the instrument will not be too high for convenient
and accurate observation.

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 189

Another framework for the purpose of supporting deflection
apparatus under the points where measurements of deflection
are to be taken is also necessary. In order that the movements
of the observers upon the observation platform may not jar the
deflection apparatus, the two frameworks must be built indepen-
dently of each other. In all the tests which have been made, up
to the present date, these deflection frames have stood on the floor
and have been braced from one to the other in order to make a
comparatively rigid framework. Fig. 12 shows scaffolding and
deflection frames for the Franks test. An objection to this method
of measuring deflections is that changes of humidity are likely

FIG. 12. — SCAFFOLDING AND DEFLECTION FRAMES, FRANKS BUILDING.

to change the length of the wooden posts used, and it is quite
probable that an improvement could be made in the form of this
frame. An arrangement which has been suggested consists of
steel I-beams supported directly by the columns and carrying
other steel framework on which can be placed the deflection
apparatus. This would give more nearly a self-contained con-
struction, and the changes of humidity and temperature would
not change the deflection readings, except as the length of column
between the platform thus built up and the floor above is changed.
Equipment. — The equipment will necessarily consist of the
following: cutting and drilling tools, portable lights for throwing

Digitized by

Google

190 Slater on Testing Reinforced Concrete Buildings.

light into observation holes, note books and facilities for doing
drafting and for reducing data.

The cutting and drilling tools are sufficiently described in
other paragraphs.

Some kind of a portable light is a necessity, as gauge lines are
often located in dark comers and as observations may be taken at
any hour of the day or night. The light shown in Fig. 20 is a

PIG. 13. — PHOTOGRAPH SHOWING VARIATION IN HEIGHT OP GAUGE UNES,
TURNER-CARTER BUILDING.

hunter's acetylene light and is quite satisfactory. The light is
attached to the forehead and may be thrown in various directions
according to the setting of the clamp attachment. The acetylene
tank may be attached to the belt or carried in the pocket.

Loose leaf note books should be provided in which the sheets
are as large as can be conveniently handled and filed. The forms for
record shown in Fig. 29 are very conveniently ruled in hectograph

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 191

ink and copied by means of a hectograph. Printed forms might be
used, but so many differences in detail are made to correspond
with the particular test in question that this would not be advisable
as too few sheets of a single form would be required to justify the
expense of having them printed.

For the most efficient work in computing results and making
sketches for records, it is important that an adequate place be
provided where some privacy may be had, where benches and
drafting tables may be used and where instruments and other
equipment may be kept. Fig. 14 shows the temporary office
which was provided in the Turner-Carter Building test. This

PIG. 14. — TEMPORARY OFFICE, TURNER-CARTER BUILDING.

is one of the portable office shanties which the Company moves
to places where work is being done. The photograph shows the
interior of the office with the observers and recorders at work
reducing the data of the test. This added equipment will add
only slightly to the cost of the test and very greatly to the efficiency
of the work. Special attention is called to it because there is a
tendency to neglect this part and to think of it as only a secondary
matter, whereas it should be considered as one of the most impor-
tant pieces of equipment.

Summary of Test Data. — ^A summary of the main features
of the building tests discussed in this paper is presented in Table I,
as it is believed that the information given there will be of assist-

Digitized by

Google

192 Slater on Tbbtino Reinforced Concrete

t. I I S • I *

BUILDXNGB.

o
Q

<3I

£1

go 2

iril

>* 00 I I »«

0909ClCOe« 09^09

S S S S 3 S S

§ § i § ill

§ i § g § § g §

11?

04 lO O 40 ^
<«' CO Q «0 <«'

iH ^ .H 1-1

100.0

1.5

10.0

2850
1800
1070
1959
1690

1270

440

2450

II i

1 1

pqpQCf^cS mPQra

£ pq

II
II

.Si
&

•8

s -s

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 193

ance in the efficient planning of and preparation for such a test.
The foUowing notes are in explanation of data given in this table:

The figures giving area of test show the total area of the floor
covered and do not count any area twice even though loaded
twice as was done in the WenaJden Building test. It does include
the area of separate singly panel tests which were made in the
Wenalden and Franks tests.

The maximum test load in lb. per sq. ft. is given in the column
under that caption. In some cases this was over only a part of
the test area. The proportional parts of the test area having the
maximum load applied were as foUows: Wenalden 80 per cent,
Powers 50 per cent, Franks 40 per cent, all others 100 per cent.

The column giving the amount of load handled includes the
rehandling due to change of position of loads. The proportionate
parts of the loads rehandled in this way were Wenalden 40 per cent.
Powers 60 per cent, Franks 80 per cent. In all the other tests no
load was rehandled.

The column giving the number of observers includes only
those reading deformations. In the Wenalden and Powers tests
another observer took deflection readings. In the Powers test
and the Barr tests, almost all the deformation readings were taken
by each of two observers, giving a larger number of gauge lines
per observer than in the other tests.

III. Instruments and Observations.

Exiensometers. — ^In the past the great obstacle to the measure-
ment of deformations in building tests has been the difficulty of
attaching the measuring instruments to either the reinforcement
or the concrete on the flat surface of a floor, and recent tests show
the necessity of making measurements of reinforcement deforma-
tion directly on the reinforcement. A satisfactory method of
accomplishing this has been provided by the introduction of the
extensometer invented by H. C. Berry of the University of Penn-
sylvania and adapted to this work by improvements made at the
University of Illinois. This instrument is similar in some respects
to the strain gauge designed and used as long ago as 1888 by
James E. Howard, then Engineer of Tests at Watertown Arsenal,

The great value of this instrument in building tests lies in the
following facts: (a) Its use makes it possible to take measurements

Digitized by

Google

194 Slater on Testing Reinforced Concrete Buildinos.

directly upon the surface of the reinforcement and concrete.
(6) With its use there is no apparatus left in place to be damaged
or disturbed during loading, (c) Due to the fact that it is portable,
measurements may be taken in a large number of places with a
single instrument. Measurements have been taken at as many

-S^et^ Lirrm.

Ecc&ntnc hole Central hole

Finishinq
tool,

FIG. 15. — UNIVERSITY OF ILLINOIS EXTENSOMETER.

as 104 points in a single test. This would call for an outlay of
from 11200 to $2500 for instruments if fixed instruments were used.
Fig. 15 shows the Illinois extensometer in its present form.
Any movement of the point B due to a change in the length of the
gauge line is transmitted to the Ames gauge through vertical
movement of point C, by means of the leg BD and the arm DC
pivoted at D. The Ames gauge is sensitive to a movement at C
of .0001 in. The ratio of the length CD to the length BD is

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 195

approximately 5 and the Ames gauge is thus sensitive to a move-
ment at B of .00002 in. (.0001 rD.-r-S). However, this must not
be taken to mean that the extensometer possesses this degree of
accuracy in measuring stresses, since some movement of the point
of the leg at B is certain to result from variation in the handling
of the instrument.

To obtain the exact ratio between movements at points B
and C the instrument is calibrated by means of a Brown and Sharpe
screw micrometer. For known movements of the point B readings
of the Ames gauge are taken and a calibration curve plotted for the
entire range of the instrument.

The first instrument of this type built by the Engineering

FIG. 16. — BERRY EXTENSOMETER.

Experiment Station of the University of Illinois was made by
arrangement with Mr. Berry for the Deere and Webber test.
It was designed by H. F. Moore and A. E. Lord, and was like the
instrument in use at present except that it had a 15-in. gauge
length and was made entirely of steel. Later on in making the
instrument for general use aluminum was substituted for steel in
order to reduce its weight and the gauge length was made variable
from 6 to 11 in. Since then several minor changes have been
made. The legs have been made stiflfer in order to reduce the
error due to unconsciously applied longitudinal thrust and the
points have been made sharper in order to reduce the pressure
required in seating the instrument. These improvements have
reduced the probable error of observation considerably.

Digitized by

Google

196 Slater on Testing Reinforced Concrete Buildings.

The extensometer loaned by Mr. Berry to the University of
Illinois in 1910 and used in the Deere and Webber test is shown m
Kg. 16. It differs from the Illinois instrument in that the move-
ment of the multiplying arm is measured by means of a screw
micrometer instead of the Ames gauge head, the point of contact
of the micrometer plimger and the lever arm being determined
by means of a telephone apparatus. The screw micrometer and
the frame of the extensometer are insulated from each other and
are connected with the poles of a small battery by means of copper
wires. A contact between the plimger of the screw micrometer and
the multiplying lever completes the circuit and the current set up
produces a vibration of the diaphragm of the telephone apparatus

PIG. 17. — LATEST TYPE OF BERRT EXTENSOMETER.

carried on the head. This method of observation is very slow and
the apparatus gets out of order very easily,

The use of the Ames gauge head instead of the screw
micrometer and telephone apparatus adopted by Mr. Moore in
the instrument used in the Deere and Webber test has greatly
facilitated the use of the extensometer. The legs of this instru-
ment also were made longer in order to adapt it to the measure-
ment of deformations of reinforcement imbedded in concrete.
Both of these modifications have later been used by Mr. Berry in
instruments which he has put on the market.

The extensometer more recently designed by Mr. Berry is
shown in Fig. 17. It is not different in principle from the one

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 197

just described. It differs from the IlUnois instrument in the
following details: (a) Instead of having a uniformly variable
gauge length ranging from 6 to 11 in. it has two fixed gauge
lengths of 2 in. and 8 in. respectively. (6) The instrument shown
here has a multiplication ratio of two between leg and arm, and
in order to make this ratio five (as in the Illinois type) it is nec-
essary to use a leg which is only one inch long. With this
arrangement the instrument can not usually be used for meas-
uring deformations in reinforcing bars, owing to their depth of
imbedment. (c) This instrument is put out with framework of
Invar steel or aluminum. While Invar steel makes the weight

FIG. 18. — BERRY EXTENSOMETER AS MODIFIED BY TRELEASE.

somewhat greater than that of the aluminum instruments, it has
the great advantage that so great dependence on an Invar steel
standard bar is avoided and the study of the temperature changes
in the reinforcement and concrete of the structure is accomplished
with greater ease.

F. J. Trelease of the Corrugated Bar Company has designed
an instrument of the Berry type and has used it in at least one
test. This instrument, shown in Fig. 18, also has as its main
feature a multiplying lever which actuates the plunger of an Ames
gauge head. The principal difference between this instrument
and the one shown in Fig. 15 is that the multiplying lever is

Digitized by

Google

198 Slater on Testing Reinforced Concrete Buildings.

vertical instead of horizontal. Results have been obtained with
it which do not differ much as to accuracy with those of the
Illinois type of instrument.

Use of Berry Extensometer. — In obtaining good results with
this extensometer, a great deal depends upon careful manipula-
tion of it. Two things which are of great importance in this
respect are (a) the preparation of the gauge holes, and (b) care
and experience in the use of the instrument.

The exact gauge length is best secured by the use of some

FIG. 19. — INSTRUMENTS AND TOOLS.

kind of gauge marker such, for instance, as is shown in Fig. 19
used for marking points where gauge holes are to be drilled. In
the work of the Illinois Engineering Experiment Station the holes
are drilled with a No. 54 drill (.055 in. in diameter). At the
beginning of the use of the Berry extensometer a number E
countersink drill (approximately 3/32 in. in diameter) was used,
but a smaller one seems to be better, because it is easier to get the
properly finished hole, and because a slight eccentricity of the
gauge holes on the reinforcing rod causes less error in manipula-

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 199

tion of the extensometer when a small drill is used. In the case
of measurements on small rods also, the 3/32 in. drill cuts away
a large percentage of the metal in the rods. Up to the present
time, for drilling these gauge holes a breast drill has been used
which is geared so that one revolution of the crank gives about
4^ revolutions of the drill. In the hands of a skilled workman
very satisfactory work can be done in this way, but where, as
quite frequently will be the case, the drilling has to be done by
persons not familiar with this kind of work, something better is
needed. A drill driven by a flexible cable attached to a small
electric motor giving a speed of rotation of 400 r. p. m. and
upwards probably would be much better. Where high carbon
steel has been encountered many drills have been broken and
even when a hole was drilled a poor job has often been the
result. After drilling the holes, the edges should be finished to
remove the burr and to round off the sharp comers. The tool
shown in Fig. 15 is designed to accomplish this purpose. Such
a tool should not be a cutting tool but rather a wearing or polish-
ing tool. A pointed magnet to remove steel dust and small frag-
ments of steel torn off in drilling would be of use. It is hard to
place too much emphasis on the proper preparation of gauge
holes.

Standard Bar. — While the careful preparation of gauge holes
is important, not less so is the use of a standard bar. The neces-
sity for it was first found in the test of the Deere and Webber
Building. Variation in temperature was sufficient to cause a
change in the length of the instrument as great in many cases as
that in the reinforcement due to the applied load. Hence it was
found necessary to make observations on an unstressed standard
bar showing any temperature changes in the length of the instru-
ment. In this test a bar of about |-in. steel was used as a stand-
ard. It was protected from rapid temperature changes by
imbedment in plaster of Paris, but kept on the floor where the
test was being made. In this way it was expected to make the
change in the length of the standard bar due to temperature
variations about equal to the change in length of the reinforce-
ment due to the same cause. To some extent this purpose was
accomplished, but as the plaster covering was thin and not very
dry the change in the standard bar must have been much more

Digitized by

Google

200 Slater on Testing Reinforced Concrete Buildings.

rapid than that in the reinforcement. In the test of the Wenal-
den Building precautions were taken to imbed the standard bar
in concrete. This practice has been kept up in tests made since
then, and in addition standard gauge lines have been established
in parts of the floor not affected by the load. These latter have
been placed both on the reinforcement and in the concrete. Fig.
20 shows the taking of an observation on a standard gauge line
in the Turner-Carter test. It can be seen that it is located in a

Fia. 20. — TAKING AN OBSERVATION ON STANDARD GAUGE LINE, TURNER-CARTER

BUILDING.

part of the floor entirely away from the loaded area. The great-
est development in the use of the standard bar has been in the
frequency of reference to it and in the development of an exact
system for the calculation of temperature corrections. It was
previously noted that a steel instrument was used in the Deere
and Webber test but that in the subsequent tests an aluminum
instrument was used. Since the coefficient of expansion for
aluminum is almost twice that for steel, it is apparent that
dependence on the standard gauge line must have been of still

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 201

greater importance in the later tests. Difficulty was found in
interpreting the notes taken on the Wenalden test, but the
greater dependence on the standard gauge line and the more
systematic use of it observed since then has very largely overcome
this difficulty. Subsequent to the completion of the last building
test participated in by the writer, standard bars of Invar steel
have been secured. Invar steel has a coefficient of expansion
only about one-sixth that of ordinary steel and its use as a stand-
ard bar makes it possible to eliminate from the results almost all
the eflfects of temperature variation. If it is desired to determine
how great are the temperature eflfects, a standard gauge line can
be placed in the floor as before in such a position as not to be
aflfected by the floor load.

It has been the practice in the more recent building t^sts for
each observer to make observations regularly on two standard
gauge lines. This is done so that one may form a check on
the other. If only one were used, a large accidental change in
the readings due for instance to sand in the gauge holes might be
mistaken for a temperature effect. If two standards are used,
any such accidental change as the above would seldom be the
same in both, and the error would be detected. An accident to
the instrument would probably cause the same change on both
standard gauge lines and the use of the two standards would not
help to detect this kind of an error. However, such errors are
usually so large as to be apparent in any standard reading and
are infrequent as compared with errors due to filling of the gauge
holes.

Deflection Instruments. — In the building tests described in
this paper deflection instruments of two types have been used,
one being that used by the Illinois Enghieering Experiment
Station and the other that used by the Corrugated Bar Company.
The former, shown in Fig. 21, consists of a screw micrometer head
of 1 in. travel, connected in tandem with an Ames gauge head
micrometer of | in. travel. The screw micrometer is designed
to cover large variations in deflections, and the Ames gauge
head, small ones. Fig. 21 shows also the method of using this
deflectometer. A plate, having a ^-in. steel ball attached, is
plastered to the surface, deflections of which are to be measured.
A |-in. bolt, which has a steel ball inserted into its upper end,

Digitized by

Google

202 Slater on Testing Reinforced Concrete Buildings.

is set into a wooden block (part of the deflection framework) in
such a way that its elevation can be adjusted to give any desired
zero reading of the deflectometer. Thus at the beginning of a
test all the zero deflection readings can be determined so that for
a considerable length of time all the change in deflection will be
shown on the Ames gauge without any change of the screw
micrometer. As larger changes take place, a second setting of

Concr^fm Floor

kVooden ■■

FIG. 21. — DEFLECTOMETER, UNIVERSITY OF ILUNOIS.

the screw micrometer will probably be necessary. The great
advantage of this instrument is the rapidity with which it can be
used. It has been found to work very satisfactorily in most
respects. A shortcoming, however, has been the lack of a revolu-
tion counter on the Ames gauge so that in case of large changes
of deflection it is possible to make an error of as much as 0.1 in.
in interpreting the readings, though this is very unlikely. This
instrument was last used in the Turner-Carter test and since then

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 203

an Ames gauge head, which has a revolution counter, has been
provided for it, so that the diflSculty here mentioned is not likely
to occur in the future.

The deflectometer used by the Corrugated Bar Company is
shown in Fig. 22 and consists of a screw micrometer depth gauge
by means of which distances for varying loads are measured
between the stationary frame and a point on the beam or floor

/ XT' '^ ' i//////P//////^//i//P/ ' ^'V^ii'f^FA c* - V ^ ■ '/

tnakh Micrometer'^

Fia. 22. — DEFLECTOMETER, CORRUGATED BAR COMPANY.

slab. It has the advantage over the one previously described
that actual distances are measured instead of changes in dis-
tance, so that if the complete reading is taken each time, there is
no possible way of misinterpreting results. It has also the advan-
tage of a much larger range of measurement. In the Barr panel
test a gross deflection of more than 3 in. took place. As the
Illinois type of deflectometer has a range of only \\ in. it could
not have been used in this test. This, however, is more than

Digitized by

Google

204 Slater on Testing Reinforced Concrete Buildings.

would often if ever occur in the test of a building. Its disadvan-
tage is that it requires a longer time to make an observation than
does the deflectometer previously described.

Observers, — Observers should be experienced in the use of the
Berry extensometer before imdertaking work on a field test. The
chances of error in the manipulation of the instrument are large
and as a rule the deformations measured are small, so that the
error is likely to be quite a large proportion of the total measure-
ment; hence it is important to reduce errors to the lowest pos-
sible limit.

Extensometer Observations. — ^If the observations at zero are
equally as good as other observations, a curve may be drawn
through all the points of any load-deformation diagram after the
test is completed, weighting the zero observations equally with
the others and the zero point shown by the most probable curve
should be used as the origin. This method involves waiting until
the completion of the test to draw these curves. It would be
better to spend much more time on the zero observations, in order
to make them reliable, than is paid to any other series. By this
means a check can be had upon the action of the structure as the
test progresses and the construction of the most probable curve
will be made more simple. To do this it is essential that several
complete series of zero observations should be taken with no
load on the floor, and it would be weU to repeat this through
considerable range of temperature to study temperature efifect
on the reinforcement and on the concrete. This study was
attempted in the Deere and Webber test, but the changes both in
instruments and in reinforcement were included in the measure-
ments and could not be separated, so no definite conclusions
could be drawn. However, with an Invar steel standard bar or
with an instrument made of Invar steel these two kinds of
changes can be separated and to some extent at least the effect
of temperature determined.

In taking an ordinary observation about five readings should
be averaged. In all of the building tests which have been made,
individual extensometer readings were recorded, but in laboratory
tests the practice of averaging the results mentally has been
adopted. This gives very satisfactory results for laboratory tests
and saves a great deal of time. It is possible that this practice

Digitized by

Google

Slater on Testing Reinforced Concrete BmLDiNGS. 205

could be adopted for field tests also. It would save time on a
test and with a good recorder the calculations could be kept up
with the observations. In the more recent building tests the
practice followed in obtaining readings for any observation has
been to reject all readings until 5 consecutive ones have been
obtained which agree within .0004 in. These five consecutive
readings then are averaged to form an observation.

Deflectometer observations have been suflSciently discussed
in the description of the deflectometer and will not be taken up
again here.

Observation of Cracks. — ^Up to very recently the observation
of cracks has been considered one of the most important features
of a test, and if carefully done it may yet add considerable to the
confidence in the results. These observations should be made
and recorded for zero load and at each increment of load. This
is one of the most tedious parts of the test, and to carry it out
faithfully requires a great deal of patience. The examination
should be minute and very thorough. One who is not familiar
with this kind of work will be likely to miss important indications
and careful supervision should be maintained over this part of
the investigation.

Special attention has been called to observation of cracks
because of incorrect ideas which apparently prevail with regard
to them. It seems to be the idea of some engineers that the type
of construction advocated by themselves is immune from cracks.
When it is remembered that plain concrete fails in tension at a
unit deformation of about .0001, it is apparent that cracks must
form when the stress in the reinforcement is such as to correspond
with this deformation, or at about 3000 lb. per sq. in. At this
stage the cracks are often too small for detection with the naked
eye, but almost always very fine cracks are foimd at stresses
ranging between 3000 and 10,000 lb. per sq. in. Thus to report
for a floor loaded to twice the designed load that no cracks were
observed is to admit one of three things, namely, that an excess
of reinforcement was used, suflScient care in taking observations
was lacking, or that not all the facts of the case were reported.

It should be borne in mind that the cracks referred to in
this pAper are often extremely minute and usually are not visible
to a casual observer. Frequently cracks have been traced with

Digitized by

Google

206 Slater on Testing Reinforced Concrete Buildings.

a lead pencil to make them distinct for the purpose of sketching,
and it seems apparent that some persons visiting the test have
mistaken these pencil marks for large cracks. At any rate reports
have been circulated as to the existence of large cracks in a test
where to the writer's personal knowledge there were none.

ACCURACY OF DEFORMATION MEASUREMENTS.

Probable Error. — The ratio of multiplication in the Berry
extensometer is not exactly equal to the ratio of the length of the
arm to the length of the leg, the error being due to the fact that
the plunger of the Ames gauge head does not always travel in a
line perpendicular to the multiplying lever. However, calcula-
tions show that this approximation results in an error in the
measurement of reinforcement stresses equal to only about one-
quarter of one per cent for an extreme case. It may be seen later
that errors of observation are large enough in proportion that this
error can be neglected.

In forming a basis for a conclusion as to the accuracy of the
figures given out as results of tests, use has been made of the check
readings taken by two observers on the same gauge lines and of
calculated probable error of the means of five readings. While
it is possible to calculate with some accuracy the probable error
of replacing the instrument on the same gauge line time after
time at one sitting, it is very difiScult to determine the error caused
by gradually cramping the quarters of the observer as the loading
material piles up. A determination of errors based on independent
checking by a second observer should be expected to eliminate
to a large extent errors of all kinds and the greatest dependence
should be placed on this kind of results.

In the test of the Powers Building most of the observations
taken were checked by a second observer and some of the results
are shown in the load stress curves of Fig. 23. The values shown
in solid circles were observed by F. J. Trelease and those in open
circles, by the writer. The zero reading for the latter is in each
case at a load of 50 lb. per sq. ft., and in order to make a direct
comparison of results, all these curves must be set over so that
their zeros coincide with the stress values at 50 lb. per sq. ft. of
Mr. Trelease's curves. Having made this correction the average
variation between all the comparable points is about 670 lb. per

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 207

sq. in. (.0000223 unit deformation), which amounts to a probable
error of approximately ±340 lb. per sq. in. (±.000011 unit
deformation).

Fig. 24 shows the results of a series' of measurements taken in
the same way on the upper and lower surfaces of a 4 x 4 in. timber
beam loaded with sacks of sand on a 12-ft. span. The points in
open circles represent measurements on the top surface and those
in crosses on the bottom surface. Determined in the same way,

S S S S ^

lo lo ^ "o o

Si'ee/ Sfress mlh per S(f /n.

FIG. 23. — LOAD-STRESS DIAGRAMS OF TWO OBSERVERS, POWERS BUILDING.

these measurements show an average probable error of approxi-
mately =*= .000017 unit deformation. As previously stated, these
check measurements must be taken to give results more applicable
than calculations of probable error of the mean of a group of
readings. However, it may be expected that where an increase
in accuracy of setting the instrument is found, a decrease in error
due to cramped quarters, etc., will also be found. In Fig. 25 is
given a curve which shows for each of four building tests the
probable error of the average of five readings. Each plotted point

Digitized by

Google

208 Slater on Testing Reinforced Concrete Buildings.

is the average of the probable errors calculated for six different
gauge lines at a given load. What this diagram may be expected
to show is the improvement in results with increased experience
rather than the actual value of the probable error. The marked
improvement in results shown here is due in part to increased
skill in the observer and in part to improvement in the instrument
itself. Fig. 26 gives a curve showing deformations in a bottom bar
of the Barr test panel as shown in the sketch. The points shown

ZOO AOO 600

L oad in Pounds

FIG. 24. — LOAD-DEFORMATION DIAGRAMS OF TW'O OBSERVERS; TEST OP A
4 X 4-IN. TIMBER BEAM.

as open circles are for a load of 590 lb. per sq. ft. and solid circles
are for a load of 615 lb. per sq. ft. This is the best curve the
writer has been able to obtain on any building test and it can not
be taken as representative, but rather to illustrate what may be
obtained under the best conditions. The regularly varying differ-
ences for a small difference of loads indicate that the stresses
must^have been determined correctly within a very small range.
A study of probable error was made in the Turner-Carter
test by the use of a series of 100 observations taken by each of the

Digitized by

Google

Slater on Testing Reinpobced Concbetb Buildinos. 209

^/OOO

eoo

^ 400

1
1

—

Hi
Is

OrvfQr

3
of Tests

PIG. 25. — AVERAGE PROBABLE ERROR FROM FOUR BUILDING TESTS.

JO0O5
\.0003

\jOOOB
^.0003

Load'd/L '^ It per t "cf Ft-

iK^ 7^t3g0Z \pffr S(^/- P"

1§ % I ^ 8"

Distance in Inches from Edge of Beam

FIG. 26. — DEFORMATION ALONG BOTTOM REINFORCING BAR, BAR PANEL TEST.

Digitized by

Google

210 Slater on Testing Reinforced Concrete Buildings.

two observers on two gauge lines selected as likely to give the most
and the least accurate results. The results of this study are given
in Table II.

TABLE II — ^Probable Error of the Average op Five Consecutive

Readings.

Obsenrer

Gauge Line.

Average.

Unit deformation

/ H. F. Moon
\ W. A. BUter

/ H. F. Moore
\W.A.SUt6r

.00000687
.0000043

206
180

.0000106
.000014

818
436

.00000873

Stren in reinf oroement in lb. periq. in .

.0000001

262
282

While these measurements were not all on reinforcement,
the probable error has been reduced to terms of stress in reinforce-
ment for convenience of interpretation. It is very interesting
to note that the average probable error of =^282 lb. per sq. in.
agrees very weU with that for the Turner-Carter test as shown in
Fig. 25. The same observer took the data in both cases, but the
data for the value shown in Fig. 25 are taken directly from the
records of the test and r^resent the condition on six typical
gauge lines. The method of obtaining the values given in Table
II is explained above.

From the data in hand it seems safe to conclude that for
ordinary conditions stresses in reuiforcement can be measured to
the nearest 1000 lb. per sq. in., though in the past there have been
some glaring failures to obtain as great a degree of accuracy as this.
The advantage of further increase in accuracy of results lies in the
determination of the relation of parts of the structure.

Effect of Changes in Temperature on Accuracy of ResvUs, —
Changes of temperature will give measureable changes of length
in reinforcement, in concrete and in instruments made of ordinary
materials. In most of the building tests corrections have been
made for the changes in the instrument due to changes in tempera-
ture by means of observations on standard unstressed gauge
lines chosen to represent as nearly as possible the conditions of
the reinforcement and the concrete in the part of the structure
tested. The method of calculating this correction will be described

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 211

below. It is there mentioned that in distributing the corrections
found by reference to the standard bar, a linear variation from the
time of one standard observation to the time of the next standard
observation was assumed. Some observations have been made
to determine the correctness of this assumption.

To determine the amoimt of change in length of an aluminum
extensometer covered and imcovered, a test was made in which
the two instruments were suddenly exposed to a change of tempera-
ture of 60 deg. F. A covering which consisted of a double layer
of rather heavy felt protected one of the instruments from too sud-

005\

.00/

(a) /nstrument

f^) /nsfrument-
Profcctod

FIQ, 27.-

/O 20 30 40

Length of Exposure //? Minutes

-DIAGRAM SHOWING CHANGE IN LENGTH OF INBTRUIIENTS DUE TO
CHANGE IN TEMPERATURE.

den a change in temperature. The other instrument was entirely
unprotected. The results of this test are shown in Fig. 27 with
the change of length of the instrument plotted as ordinates against
time as abscissas. For these measurements a standard bar of
Invar steel was used. The coeflScient of expansion of this being
very small, the change of length measured must have been almost
entirely that in the instrument. The curve shows that for an
instrument not insulated from temperature changes only about
five minutes is required for the instrument to come to the tempera-
ture of the air. For the insulated instrument about 20 minutes
was required. This may be interpreted to mean that if an unpro-

Digitized by

Google

212 Slater on Testing Reinforced Concrete Buildings.

tected instrument is used, readings on the standard bar should
not be more than five minutes apart. With an instrument pro-
tected as was this one, intervals of 20 minutes would not be too
much. The amount of change for the case shown here is extreme
as the instrument was suddenly exposed to a change in tempera-
ture of about 60 deg. F. This range would seldom be found,
and the length of time required to make the change for a smaller
difference of temperature may be less but probably would not
vary much for other ranges of temperature. It may he concluded

jOOSO

^ jOO/C
X)OOS

/DOCO

flX"

(6)
\

' «^

o /o ao 3o ^o so
Time /n Minutes.

Fia. 28. — DIAGRAM SHOWING CHANGE IN LENGTH OF A STEEL BAR DUE TO
CHANGE IN TEMPERATURE.

(a) |-in. square bar exposed.
(6) |-in. round bar embedded in concrete.

that the method used for distributing the correction is justifiable,
since the instrument was protected from sudden change of tempera-
ture and the observations on standard bars were usually at inter-
vals not greater than 20 minutes.

Temperature Effect on Reinforcetnent. — The above test shows
the effect of change in temperature on the instrument. Another
test was made to determine the effect of change in temperature on
reinforcement imbedded in concrete and also exposed to the air.
A f-in. square bar of steel entirely unprotected from temperature

Digitized by

Google

Slater on Testing Reinforced Concrete Buildij^gs. 213

changes and a f-in. round bar imbedded in 1 in. of concrete were
exposed to a sudden change of temperature of about 43 deg. F.
Measurements were taken on a 6-in. gauge length of each bar at
very short intervals of time. The results are shown in Fig. 28.
The results of this single test must be used with caution as the
total measurements were very small and a small error would
show up very plainly. However, the curve for the imbedded bar
agrees in its general characteristics with some of the results
obtained by Professor Woolson on '* Effect of Heat on Concrete"
reported in 1907.* The test indicates that for this range of tem-
perature rather rapid changes may be found in the reinforcement,
corresponding with stresses of about 9000 lb. per sq. in. and 3000
lb. per sq. in. respectively for exposed reinforcement and that
protected as in this case. The range of temperature is extreme
and the size of bars smaller than is often found in floor con-
struction, therefore the results found in tests would probably be
less extreme. However, this indicates the necessity of attempting
to eliminate from the results of the test the effect of tempera-
ture changes, especially if the stresses measured are small.

IV. Records and Calculations.

Since the beginning of the use of the Berry extensometer
for testing purposes, as much development has been made in the
keeping of notes as in the use of the instrument. Because of a
lack of completeness of notes the advantages of the use of the
standard bar were not fully realized for some time. Only after
the method of keeping notes had been highly systematized was
it possible to properly make the corrections which observations
on the standard bars indicated- should be made. During the
time of placing an increment of load the recorder will have con-
siderable time in which to be working up results of the series of
observations taken at the previous increment of load, and as the
method of making these calculations is quite intricate, a man is
required for this work who has ability to do more than merely
record. It is important that calculations should be kept up as
the work progresses, because it can be done with less labor then
than at any other time and because it will be of value to know
as the test progresses what results are being secured.

• See Proceedtnga, Vol. VII. Am. Soc. for Test. Mala.

Digitized by

Google

214 Sla-bpr on Testing Reinforced Concrete Buildings.

RECORDS.

It is very important on account of the great number of
observations taken (about 12,000 in the Turner-Carter test)
that all records be arranged systematically. The following
points are mentioned as being important in this connection,
(a) In the field test individual readings should be recorded and
their average used as a single observation. The proposed abridg-
ment of this procedure (see p. 204) should be considered as a sug-
gestion for later development. (6) Recording readings in the
order of their size will assist the recorder in obtaining the correct
readings and in rapidly obtaining the average, (c) The exact
sequence of observations should be maintained in the records
as the calculation of corrections depends largely on this.

Fig. 29 shows a form for the recording of original readings
and the results calculated from them.

calculations.

The calculations of corrections and applying them to the
results makes the reduction of data rather intricate. This work
has been reduced to a definite form shown in Fig. 30. In this
form the zero length of instrument (see definition, p. 172) is assumed
as correct and is used as a standard of reference. The correc-
tions are distributed among the gauge lines as though the change
in the length were a linear function of the time from one standard
bar observation to the next one. These assumptions do not
entirely accord with the facts but have been satisfactory as a
working basis. Any other standard bar observation than the
zero length would do as well for a standard of reference except
for matters of convenience.

V. Cost of the Tests.

An attempt was made to get information by which the cost
of the tests could be estimated, but it is found that from the
data on hand no finely drawn conclusions are warranted. The
costs of the tests enumerated here range from about $50 to as
much as $2000, depending on the nature of the test and the expenses
for railroad fare, hotel bills and pay for expert assistance. In
the case of $50 the cost is only that in excess of what would have

Digitized by

Google

Slater on Testing Reinforced Concrete Buildinos. 215

Digitized by

Google

216 Slater on Testing Reinforced Concrete Buildings.

«Vj

P007

^<^

1:^ ^

^H5

"1^

v%'

2^^

/fao J0007 po9a

't>

«^

vj ?

*> N>ft

."8 3

•5 a

«^

9
■^

(^ fiooj 3H/7 /(Ufr

O
§

o
o

Digitized by

Google

Slater on Testing Reinforced Concrete Buildings. 217

been necessary for the acceptance test of a building. It does
not include the salaries of the testing engineers and as there were
no hotel expenses this cost is perhaps $150 less than what could
be ordinarily expected for such a test. The data in hand indi-
cate that a test as extensive as several in this series have been
would be likely to cost from $1500 to $2000^ but that one of the
type represented by the Carleton test may be made at an expense
slightly above that of the acceptance test so frequently required.

For a very slight additional cost, measurements of stresses
in a building floor may be made at points of especial interest
during the progress of the load test which is often required as a
condition of acceptance.

The stage has been reached in the investigation of reinforced
concrete where building tests may be expected to contribute
information of great value to the designer and builder in rein-
forced concrete. The main feature of such tests should be the
measurement of stresses, but information as to the location and
size of cracks will be of great value in checking the results if the
examination for cracks is conducted with sufficient care and
minuteness. There is need for increasing as much as possible
the accuracy of deformation measurements and experience in the
use of the instrument is gradually accomplishing this. All the
confirmatory evidence possible on the correctness of results should
be obtained.

Digitized by

Google

THE DESIGN OF CONCRETE FLAT SLABS.*

By F. J. TRELEASE.f

The rapid introduction and development of flat slab floors
of reinforced concrete is sufficient evidence of the great importance
of this type of construction. In fact, the demands of owners
and builders for flat slabs are so insistent, that in spite of the lack
of reliable methods of design, many acres of such floors are built
every year. The insistence of this demand has led many engineers
to advance theories of the structural action of flat slabs, and these
theories yield perhaps the most striking contrasts and disagree-
ments to be found in modem engineering practice. The reason
for thib is largely because the flat slab is an extreme example of a
redundant structure, and its mathematical analysis has so far
been based upon certain arbitrary assumptions which vary in the
different analyses. The differences in these assmnptions are so
great that the practical application of the theories founded upon
them to flat slabs of reinforced concrete yield results varying by
about four hundred per cent. This variation leads one to class
the flat slab as beyond the range of pure analysis and one must
look to experimental engineering for a satisfactory solution of the
problem.

The usual load tests of completed structiu*es may be dismissed
at once as being entirely inadequate as a basis of design. In such
tests usually but one panel of a structure is loaded, which does not
give maximum stress conditions in a flat slab floor. Even if several
panels be loaded and the test carried to destruction it will at best
only roughly indicate the stresses at the weakest point of the
structure under the scheme of loading employed and cannot give
any information as to the economy of the design.

A very recent and more adequate form of test on completed
structures is that in which several panels of a building are loaded,
not necessarily to destruction, and in which the actual elastic
deformations of both reinforcement and concrete are measured

* Advance Review of a Thesis presented to Washington University for the degree of Civil
Engineer.

t Engineer, Corrugated Bar Company, Buffalo. N. Y«

(218)

Digitized by

Google

Trelease on Design of Concrete Flat Slabs. 219

by extensometers. The results obtained in such a test would
form a satisfactory basis for the design of similar structures if one
could eliminate from the results the effects of tension in the con-
crete, arch action, annular slab action and many other indeter-
minate factors, which enter largely into all tests of reinforced
concrete. These effects are so indeterminate that they cannot
be satisfactorily eliminated, thus rendering results of tests on
completed structures more or less imsatisfactory as a basis of
design. The average engineer hesitates to coimt on these factors
in designing and even if he had no personal objections to their
use, none of the building laws or regulations will permit of any
allowance being made for them.

It would seem, then, that for satisfactory solution of the
problem there must be obtained an empirical analysis of the struc-
tural action of flat plates derived from experiments on plates
of homogeneous material.

Such an analysis could be then used in the design of flat slabs
of reinforced concrete, using any desired combination of unit
stresses ^nd making such allowances for tension in the concrete,
etc., that personal judgment or various building regulations might
dictate.

It is the purpose of this paper to describe and give the results
of such an empirical analysis based on experiments conducted
by the author in the Research Department of the Corrugated
Bar Company, under the general direction of Mr. A. E. Lindau,
Chief Engineer. This work was started early in 1910, and has
continued almost iminterruptedly to the present time.

A very interesting and exceedingly simple little experiment
was made on a sheet of heavy cardboard fastened over twelve
spools representing columns. By pressing the fingers on this
little model at various points much was learned as to the general
action of flat plates. For instance, one could press upon the center
of several panels and note that the surface of the model was convex
upwards at right angles to the lines joining the columns and form-
ing the sides of the panels, showing that tension in the top face
existed at that point, although none of the systems of reinforce-
ment then in use provided resistance to these stresses.

The work most productive of results was a series of experi-
ments on rubber models of flat slab floors. Rubber was chosen

Digitized by

Google

220 Treleasb on Design op Concrete Flat Slabs.

for the models as best fulfilling the requirements of reasonable
homogeneity, low modulus of elasticity, and the ease with which
all its physical properties could be determined in the laboratory.
The models used consisted essentially of a plate of pure gum rub-
ber, fixed so as to form the top of an air-tight box containing the
proper number of 1 J in. round columns to divide it into nine panels,
each 8 in. square.

The method adopted for applying load to the model was
rather unique and extremely satisfactory, as it not only permitted
the intensity of the load to be easily and accurately read, but

FIG. 1. — BOX AND ABPIRATOR FOR RUBBER MODEL OF FLAT SLAB FLOOR.

insured absolute uniformity of distribution, and at the same time
left the entire upper surface of the slab unobstructed and free
for observations and measurements. A partial vacuum was
formed in the box, thus obtaining on the face of the plate the pres-
sure of the atmosphere which was read by a simple U-tube manom-
eter. The box and aspirator are shown in Fig. 1.

In the first series of experiments the rubber plate used was
0.34 in. thick. A stress strain diagram obtained from a strip
cut from this plate is shown in Fig. 2. At first the scope of \his
series was limited to the measurement of the shape of the elastic
surface of the plate under stress. With this end in view, deflection

Digitized by

Google

Trblease on Design of Concrete Flat Slabs.

221

readings were taken at numerous points in the various panels.
In some cases these were taken at points only one-twentieth of
the span apart.

The values of these readings were then averaged, grouping
those which through symmetry should be alike, and the results
plotted. Equi-deflection lines were then drawn, giving contour

■~"

"~"

"""

~~"

"""

""■

^iU

^ 6 a /o /e M /6 /3 so K Z^ ,» ZS ^30
FIG. 2. — STRESS STRAIN DIAGRAM FOR RUBBER.

maps of the surface. Fig. 3 shows such a map for one-quarter
of the model, while Fig. 4 shows the lines of equi-deflection for
the central panel. The approximate location of the lines of
inflection for imaginary beams radiating from the column center
have been plotted in Fig. 5.

Many interesting conclusions can be drawn from the deflection
maps. They show at once the general nature and intensity of the

Digitized by

Google

222 Trelease on Design op Concrete Flat Slabs.

stresses existing in the plate. They show beyond doubt that
tension exists in the upper face of the slab at right angles to the
lines joining the columns and forming the sides of the panel, being
a maximum at the colunms and decreasing to a minimum at the
mid-point, although even this minimum value of the tension in the
top face is approximately equal to the tension in the bottom face
at the center of the panel. Inspection of Fig. 6 will make this
more clear. The top curve is a section along the side of a panel

WQ. 3. — CONTOUR MAP OP ONE-FOURTH OF MODEL.

and the lower curve a section through the middle of the panel
parallel to one side. From the radii of curvature of these curves,
it will be seen that the maximum stress at the side occurs over the
colmnn, while the section through the middle shows that the stress
is of practically equal intensity at the center of the panel and at
the mid-point of the panel edge. Tension in the top surface
at the edge of the panel has never been provided for in the rein-
forcing of flat slabs, although the need for it has been clearly
pointed out time and again by the formation of cracks in the

Digitized by

Google

Tbblease on Design of Concbete Flat Slabs. 223

FIG. 4. — CONTOUR MAP OF INTERIOR PANEL.

FIG. 5. — USE OF INFLECTION FOR IMAGINARY RADIAL BEAMS.

Digitized by

Google

224 Trelease on Design of Concrete Flat Slabs.

concrete. These cracks, which run from column to cokmm
along the sides of the panel, have formed in nearly every multiple
panel load-test of a reinforced concrete flat slab which has been
made, and have even formed in several floors under their dead
weight only. It was hoped that the radii of curvature of the
elastic surface could be accurately measured and the bending
moments deduced from them. It was foimd, however, that
results so obtained were too indefinite to be considered reliable.
It was evident that to obtain exact information as to the
distribution and magnitude of stresses existing in the plate, it

5ecf^/o/7 a^ ^ of Co/(//7?/?^.

^ecf/on fhroi/^^ d of Panel

FIG. 6. — SECTIONS AT EDGE AND CENTER OP INTERIOR PANEL.

would be necessary to measure the actual elastic deformation of
the material caused by these stresses. This was attempted in
many ways; by measuring distortion of squares ruled on the
slab; by measuring cracks formed in plaster of paris coatings and
lines, and by various forms of extensometers. After trying many
devices, it was decided to use a microscope fitted with an ocular
micrometer and to measure with it the deformations occurring
over a gauge length 0.5 in. long. The ocular micrometer is a
disc of glass, engraved with a fine scale, which is inserted in the eye
piece. The disc lies in the plane of the image formed so that the
engraved lines appear to be upon the image itself, and its length

Digitized by

Google

Trelease on Design of Concrete Flat Slabs. 225

(

,4f

'i.

A^^J

mM^^^^*

>- Jl

-C ^-

iiip^

fcj^i.^

1^^^^^^^^

^^^^^^H

'^^^^^^^^^^^^

^^^^^H

•^^^^^^^^^h

^>^ ^^^

• ^^^^^^F

^^^^^^

^^^^j

^^^^^1

lilu

^^^^^^I^^^^^H

ria. 7. — MODEL AND EXTEN80METERS.

Digitized by

Google

226 Tbelease on Design of Concrete Flat Slabs.

can be easily read. The absolute value of one division of the scale
depends upon the magnification used, and was approximately
0.00048 in. in these experiments. The total field was about one-
fortieth of an inch.

Several devices were tried for bringing the deformations
occurring over the gauge length into the field of the micrometer.
At first a strip of heavy tinfoil was fastened at each end of the
gauge length by passing a fine needle through it into the rubber.
The pointed ends of these strips nearly touched at the middle of
the gauge length and the distance between them was measured
under no load, and again imder the various loads employed. This
arrangement of tinfoil strips was not satisfactory, as readings
taken with it were not consistent with each other, caused probably
by slight play of the strips around the needles, or, perhaps, by
the strips themselves buckling as the gauge length shortened.

After many trials, the following method was adopted and
foimd entirely satisfactory. A small piece of very fine needle
was placed point up in the rubber at one end of the gauge length
and at the other end was placed, point down, a piece of needle
with the exposed end ground to a triangular shape and highly
polished. A small strip of brass was gfoimd to a knife edge at one
end, and at the other a small conical depression was made. The
point of the first needle entered this depression, and the knife
edge was very close to the triangular end of the other needle.
The distance from the end of the strip to one point of the triangle
was read very easily, and the arrangement gave per^pct satisfac-
tion. Fig. 7 shows a general and a detail view of the extensometers
in place.

In this first series, these extensometers were arranged on
lines radiating from the columns so that the readings both along
the lines and perpendicular to them were obtained. Full results
of these readings are not given, because it was found that the
deformation at a given point increased more rapidly than the
loads, indicating catenary action or pure tension throughout the
cross-section of the plate. If such action existed, the tensile
deformations read in the top fibers of the plate would be higher
than the true values caused by bending alone and likewise the
compressive deformations of the top fibers would be less than they
should be. To test for this, short colmnns were clamped on the

Digitized by

Google

Trelease on Design of Concrete Flat Slabs. 227

top of the slab directly over those on the under side and the load
reversed. Zero readings for the reversed load were taken under a
pressure equaling the dead weight of the slab. Fig. 8 shows read-
ings obtained in this way and under direct load, for points along
the line forming the side of the panel. The values of the deforma-
tion due to bending moment alone are the arithmetical averages
of the direct and reversed readings, and the true zero line has been

\11

\so

\r^

^ZO

'%/3

^«M

.«/

rJJ

522

■^

;J5

\ ^

Si o

-.,

Ty<6

?-r;

(

►

(

2 '4 e

FIQ. 8. — DEFORMATIONS UNDER DIRECT AND REVERSED LOADS.

plotted showing the amount of catenary action existing. This
method of load reversal was too tedious to be adopted and it was
decided to use a thicker plate in the next series of experiments.

The results obtained in the first series of experiments have
been discarded in favor of those in the second series and the first
series has been regarded as merely preliminary and as having
served its purpose in enabling the many mechanical difficulties
to be overcome and in indicating many points to be covered.

Digitized by

Google

228 Trelease on Design of Concbbte Flat Slabs.

In the second series of tests the apparatus used was the same
as that employed in the first series, except that the rubber plate
was 0.485 in. think. This greater thickness was used because
it was thought that the effect of catenary action would be less
on a heavy plate than the lighter one used in the first series. The
correctness of this assumption was later proved by a series of
readings of deformations at various points under direct loads

6 3 /o /jt /^ /3 /a 2o rz ^

FIG. 9. — propobhonalitt of btress to load for second model.

equivalent to from 1 to 6 in. of water which showed proportion-
ality of deformation to load. The results of these readings are
plotted in Fig. 9. The lack of catenary action was also confirmed
by reversing the load and obtaining curves identical to those
obtained under direct load.

Deflection readings were taken at points over the interior
panel spaced one-tenth of the span apart. Fig. 10 is a plan of
the interior panel upon which have been plotted contour lines

Digitized by

Google

Trelease on Design of Concrete Flat Slabs. 229

FIG. 10. — CONTOUR MAP OP INTERIOB PANEL.

FEG. 11. — UKEB OF INFLECTION FOB STRIPS PARALLEL TO SIDES OF PANEL.

Digitized by

Google

230 Trelease on Design of Concrete Flat Slabs.

representing elastic deflections of 0.01 in. caused by a load equiva-
lent to 5 in. of water. Sections of the surface so defined were
taken parallel to the sides of the panel and the points of inflec-
tion plotted in Fig. 11. The lines joining these are lines of inflec-
tion for imaginary beams or strips parallel to the sides of the
panel. These lines agree very well with the results obtained
later from deformation readings.

Fig. 12 is an exaggerated sketch of the elastic surface upon
which have been marked coefficients for deflection. To apply
these to concrete flat slabs reinforced parallel to the sides of the
panels a strip of unit width is assumed along the edge of the panel,

FIG. 12. — DIAGRAM OF ELASTIC SURFACE AND DEFLECTION COEFFICIENTS

and another at right angles to this along the center line of the
panel. Knowing the total load per square foot on the panel,
as well as the eflfective depths and steel percentages of each strip,
it is an easy matter to obtain the deflections. Instead of work-
ing out the values of the moment of inertia of the strips one may
follow the method outHned by Mr. Eli White in the Engineering
Record of November 9, 1907, and elaborate<l by Mr. G. F. Dodge
in his Diagrams for Designing Reinforced Concrete Structures.
To simplify the computations the strips may first be treated as
simply supported beams, and the deflections so obtained multiplied
by the ratio of the deflection coefficients.

The results of the first series of tests indicated that tensile
reinforcement would have to be provided in the top of the slab

Digitized by

Google

Trelease on Design of Concrete Flat Slabs. 231

at right angles to the lines fonning the sides of the panel, and
it was decided that a two-way system of reinforcement would be
most adequate, both in providing this reinforcement and in tak-
ing care of the stresses existing in other parts of the construction,
and consequently the arrangement of the extensometers was as
shown in Fig. 13. It will be seen from this figure that deforma-

-.s \ ^ ^ ^r

^ ^ N ^ /-

s.,

■*

J K

\ r

^ ^5 (J ''i S <& s

PIG. 13. — ARRANGEMENT OF EXTENSOMETERS

tions were read parallel to each side of the panel at points spaced
one-eighth of the span apart, and so distributed that deformations
in other portions of the plate were by the symmetry of the con-
struction obtained from readings on these.

It was found that a load equivalent to 5 in. of water gave
deformations large enough to be easily read, and this load was
therefore adopted. The extensometers at all the points were

Digitized by

Google

232 Trelease on Design of Concrete Flat Slabs.

to
/3

k_ —

^ -

' ^_^

^ -

^ .

^ ,

"^^

J>efor/y?af/o^3 aiy^ //ne ^-/l

■^

''^^

.^^

" —

"-"^

DeA

or/na;

MU7J 4

f/or,^

//-v ^.a

■

*"**-^

— "

/>eA

'orma^

J'OW <

///JK? CC

"N,

—

^v^

=^^

Z>e/a

rma/A

0/7J <^

3^/

//re Jj

-z?

{

V ■

»v

.^ ;_

?'«^

2?e/b

r/7?af/

0/7J <i

/^^

O/fe J.

rjT

V 4

^ — '

V o

""^-

^-"^

Z^e/b

rmo//

ao? dM

OJ^l

//7^ O

■a

( .

h-*^

^,„»-

""•^

^___^

^ — .

""^^

Z?e/l

vmaA

^0/73 a

/to^^

'//7e /

■/-

■

■^

"^e/o

'-ma//

1/rj a/

'<y^ /.//?e A-'/C

*-^>^

^^^

-"^

^/c

VYTfO//

0/f3

^0/7£

Z//?e

z-z

Pia. 14. — MOMENT CURVES.

Digitized by

Google

Trelease on Design of Concrete Flat Slabs. 233

^-^m

""^->.

^-^..^

I>/j/r//)t//xv7 acrajj lf/?e ^-A

"*^

^ ,

""^

1 ^

"""^^i

r-.

D/j/r/itHz/fo/? ocroJ3 l/ne S3

" ,

^ '

.--*ij

^..^^^

D/sfr/tk/^/on acroxs A//>e

^-^^

^>-^

" '

^— -^

J)/jfr//>t//

/au7 ocrosj l/ne

^— V

/^^

^.>^

•""^

r—

►— —

/>.

fjfr/bi

///on

aero.

w l/ne £^%JL

y-^

■

I>/3/r//>t///o/? across One OO

o
°

^""^

•-"— .^

>- H

!>/

3/r/i)o

//on t

7cro33 l/ne

'/■/

-A

Z!^

" H

"— —

— — "^

""-"^

r-*— -.

/?.

'jmbi/f/o/?

acro^

^3 l//fe A'/r

^49

—

/j/r/A

u//on

f7cra.

^S l/A

V li

\ — '

' ~-

-. — ^

---'

*'*"**^

■^

x^J

/)/j/r/d<///on

oe/vj3 Z/ne /^-/f

FIG. 15. — ^DISTRIBUTION CURVES.

Digitized by

Google

234 Tkeleasb on Design op Concrete Flat Slabs.

read under no load, and again, under load; the differences giving
the elastic deformations caused by the load. The load was
applied and released time and again, as many as forty readings
being taken on some of the critical points by each of two observers.
The readings obtained at any given point agreed with one another
to within a very few divisions of the micrometer. Table I gives
the averages of all the readings taken. It will be seen upon inspec-
tion of Fig. 13 that two sets of curves may be plotted from these
readings. One set will show moment diagrams for imaginary
strips, the axes of which lie along the zero lines of the curves.
Curves of the other set show the distribution of moment among
various strips of this kind. Since a two-way system of reinforce-

Table I. — Averages of Deformation Readings.

Point.
Line.

Bos.

C. L. of
Exterior
Panel.

Tane

of
ColB.

C. L. of
Interior
Panel.

B
C
D

I
K
L

18.8

9.17
14.0

6.6

0.0
4.5

-2.26

-3.0
-2.0
-0.83
-0.67

-1.83

-6.0

-0.5
-6.5
-4.1
-3.0

0.5

-1.0

-3.0

-1.17

-0.25

-1.67

-0.6

3.82

6.5

0.33

17.33

21.25

10.33

2.54

2.0

2.67
4.25
6.33
7.0
10.61

6.0

0.5
-0.5
-0.25
-1.0
-4.0

-0.6
-1.33

-3.0

-3.76

-4.75

-«.o

-8.0

-4.03
-5.6
-6.0
-10.0
-0.66
-7.6
-3.33
-2.0
-0.75

ment was to be used, it was decided to design each set of bars
independently. For square panels the reinforcement would, of
course, be the same in each direction.

Fig. 14 shows the moment curves for strips parallel to one
side of the panels, while Fig. 15 shows distribution curves for
these same strips. These two sets of curves can be combined to
form a surface showing both moments and distribution for one
set of strips. A photograph of this surface for four panels in one
comer of the model is shown in Fig. 16.

Fig. 17 is a sketch showing the surface for the interior panel.
In this figure the ordinates from the plane A-B-C to the surface
are proportional to the deformations of the top fiber of the plate
parallel to side B-C at the point at which the ordinate intersects
the surface. Sections parallel to B-C will be the moment dia-

Digitized by

Google

Tbelease on Design of Concrete Flat Slabs. 235

grams similar to those in Fig. 14, and sections parallel to A-B
will be distribution curves similar to those shown in Fig. 15.

In Figs. 14 and 15 the ordinates to the ciu^es are shown in
terms of divisions of the micrometer. In order to co-ordinate
divisions of the micrometer with bending moment, a strip of the
rubber was cut from the plate, and the stress strain diagram shown
in Fig. 18 obtained. Both the axial and lateral deformations of
the strip were measured, from which the modulus of elasticity
was found to be about 1,000 and Poisson's ratio about .44; one

FIG. 16. — MOMENT SURFACE FORMED BY COMBINING MOMENT CURVES AND
DISTRIBUTION CURVES.

division of the micrometer corresponding to a unit elongation of
0.001.

The value of one division of the micrometer in terms of
bending moment was, however, obtained directly in the follow-
ing manner:

A strip of rubber cut from the plate was supported over two
knife edges, and loaded as shown in Fig. 19. An extensometer
exactly like those used on the plate was placed on top of the strip
between the supports. Known weights were applied, and the
accompanying deformations read. Fig. 19 shows deformations
so obtained, plotted against bending moments in in.-lb. per inch

Digitized by

Google

236 Trelease on Design of Concrete Flat Slabs.

width of strip. From this curve the moment equivalent of any
number of divisions of the micrometer can be obtained. The
unit load on the panel being known^ together with the span,
what might be termed the "gross bending moment" or M=wS^,
may be obtained. In this case the clear span S was 6.75 in. and
the load equivalent to 5 in. of water, or .1808 lb. per sq. in., so
that M=wS^ = 8.2377 in.-lb. If the net moment at any point
be ilf == wS^/z, the value of the z may be obtained from the moment
curveB in Fig. 14 and the calibration curve in Fig. 19. With S

FIQ. 17. — ^DIAGRAM OF MOMENT BURFACB FOR CONTtNUOUB FLAT PLATS
SUPPORTED AT CORNERS.

equal to the clear span from edge to edge of column heads, z »
215/D, where D is the ordinate to the moment curves in divisions
of the micrometer.

On Fig. 20 the moment expressions obtained in the above
manner have been marked at the critical points.

Having now obtained an empirical analysis of a homogeneous
plate it will be applied to flat slabs of reinforced concrete. This
may be done without hesitation because the whole history of the
development of reinforced concrete has been along such lines.
It was found that once having solved the problem of internal

Digitized by

Google

Tbelease on Design op Concketb Flat Slabs. 237

stress distribution in a concrete beam, there could be safely-
used the equations for external moments applying to homogeneous
beams, and so on.

In this empirical analysis the components in two directions
of the stresses existing in a homogeneous plate have been measured
and reinforcement will be supplied in the concrete flat slab to
resist these components where they show tensile stresses. In

•

•i

r ,

\- -

•w

•>

\/o

•

}

1 i

r~-

I ,

\ /

ia*

nac//a^

to/

V
►/7

•4

no. 18.

^/i^/j/o/fj o/' Af/cra/7?efer

FROM PLATE LOADED FOB MOMENT CALIBRATION OF BXTEN-
80METEB8.

laying out the reinforcement a departure is made from the old
diagonal or four-way system and the two-way system, Fig. 21,
employed.

One of the greatest advantages of the two-way system is
the location of tensile reinforcement in the top of the slab between
the columns, which tends to prevent the formation of cracks
extending from column to column and stiffens the whole struc-
ture, giving all the advantages of continuous construction.

Digitized by

Google

238 Tbelease on Design op Concrete Flat Slabs.

Another point is that the tensile reinforcement over the columns
is in but two layers and hence the effective depth at this critical
point is much greater than would be possible with the systems
in which four or more layers are used.

' ^x/>e/9SomeAfr

Z'/4'

/fk'

£'/■*'

^ 1

■"*'

"~"

~~'

"""

"*"

"~~

~~"

■""

"~*

}

/./

5 AO

(

\a6

\^7

\a6

\a.6

^04

\o.5

^AZ

'■w

O.A

FIQ. 19. — MOMENT CALIBRATION CURVE FOR EXTENSOMETER.

The ideal reinforcement would be such that its area at any
point be directly proportional to the ordinate of the moment
surface shown in Fig. 17, but such reinforcement is obviously
impossible in practice. The panel is, therefore, assumed to be
divided into ten strips or imaginary beams, five parallel to each

Digitized by

Google

Tbelease on Design of Concrete Flat Slabs. 239

side, and such moments assumed for their design as will give
reinforcement areas conforming most closely to the theoretical
values.

The widths of these imaginary beams together with the
moment factors applying to each are shown in Fig. 22. ^ To
avoid confusion in the drawing, the widths and moments for but

FIG. 20. — ^MOMENTS PER UNIT WU>TH DERIVED FROM EXPERIlfENTB.

one set of strips are shown, those for the other set being the same
but turned through ninety degrees. The numbers in the circles
on the diagrams are moment coefficients entering into the denomi-
nator of the moment equation. Thus, the moment at the center
of the panel in the middle strip or beam is Af = i6JjSV40, where
w is the total unit live and dead load on the panel in lb. per sq. ft.,
« is the clear span in feet, and the 40 is the moment coefficient

Digitized by

Google

240 Trelease on Design op Concrete Flat Slabs.

Digitized by

Google

Trelease on Design op Concrete Flat Slabs 241

shown in the circle at this point. M is the bending moment in
ft.-lb. per ft. of width, and is positive, requiring reinforcement
in the bottom of the slab, as indicated by the fact that the arrow
showing the direction of the reinforcement is dashed. The
moments of other points of the panel are obtained in the same
manner. The sections, at the bottom of Fig. 22 show how the
theoretical distribution curves are replaced by more practical
stepped lines. In making this change an excess of reinforcement
has been introduced at the center of the strips to provide for
single panel concentrations of loading.

Since the maximum stresses in a flat slab occur at. the col-
umns, the use of a slab of uniform thickness is wasteful of concrete
and adds mmecessary dead load. The imnecessary concrete
can be done away with, decreasing the weight and cost of the
structure and at the same time adding to its efficiency, if a rec-
tangular cap of concrete is left over the columns, extending to
about the fifth point of the span.

The general method of procedure in designing a concrete
flat slab in accordance with this method is to first find the areas
of reinforcement required in the different bands at the center
line of the panel. These areas are supplied by rods of the proper
size and spacing, every other one of which is bent up in the top
at about the quarter point and extends over into the adjacent
panels, while the remainder are straight and extend about 6 in.
past the edge of the panel. This arrangement gives the necessary
amounts of reinforcement in both the top and bottom of the slab,
except in the areas over the columns where the necessary extra
reinforcement in the top is supplied in the form of short, straight
bars, remembering that at these points the effective depth of the
slab is increased. To avoid bimching the bars in the top of the
dab, a bent bar in one panel should be in line with a straight
bar from the next panel.

The line of inflection may be safely taken at the fifth point
of the clear span, or at the quarter point of the center to center
spacing of the columns, using the higher value in every case. The
point of bend in the bent bars, and also the length of the extra
short rods in the top, will be governed by this dimension.

In the end panels the area of the reinforcement perpendicular
to the wall should be increased by the usual 20 per cent. The

Digitized by

Google

242 Trelease on Design of Concrete Flat Slabs.

-^)-r-^

■-^-^

-P^

~f ^

-^o

-^H-^

■^-^

■^v

"Ol

^ ^y

D/jfr/Jbt/f/o/7 across l//?e XX

T
_L

^/s^r//>u//o/? across Z//?e Y-y.

FIQ. 22. — ^MOMENTS PER UNIT WIDTH USED IN DESIGN.

Digitized by

Google

Trelease on Design op Concrete Flat Slabs. 243

easiest way to do this is to use round bars in the interior panels,
and square bars of the same nominal size and spacing in the end
panels. In order to facilitate the placing of the reinforcement, it
will be found necessary to provide light bars, preferably | in.
round, along the sides of the panels, so that they may be used to
support the bent bars. Of course, these spacing bars may be
figured as effective in tension over the top of the colunm, and thus
replace some of the short bars used there.

Table II gives recommended dimensions for flat slab floors
to suit various live loads and panel sizes. The table is based on
a maximimi theoretical concrete stress of 750 lb. per sq. in. The
slab thicknesses are such that the deflections under a superimposed
test load equal to once the dead load plus twice the live load
will not exceed one five-hundredth of the span for a theoretical
working stress in the reinforcement of 18,000 lb. per sq. in. Ten-
sion in the concrete, arch action, etc., all tend to reduce the actual
stresses below the theoretical values, but no building code will
permit these factors to enter into the design of concrete structures.

Flat slab designs showing smaller column head diameters,
or thinner slabs over the supports, than those given in the table
should be carefully checked for shear at the edge of the column
heads.

To illustrate the application of this new method for the
computation of flat slabs, the detailed design of a typical panel
will be given. To enable comparison of the results with those
obtained by other methods described and tabulated by Mr. Angus
B. McMillan in a paper* before the Association, a panel 20 ft.
square will be designed for a live load of 200 lb. per sq. ft., using a
steel stress of 16,000 lb. per sq. in. and a ratio of the moduli of
steel and concrete of 15.

In Table II under the above span and loading a slab thickness
of 8 in. is given, with a 2 in. cap 8 ft. square, and a column head
50 in. in diameter. The average weight of the slab, including the
cap, is 104 lb. per sq. ft.

The clear span from edge to edge of colimMi heads is 15.83 ft.,
and the "gross bending moment" or wS^ is 12X304X15.83^ or
914,150 in.-lb. per ft. of width. Referring to Fig. 22, it will be

* Proeudingg, Vol. VI, p. 248.— Ed.

Digitized by

Google

244 Trelease on Design op Concrete Flat Slabs.

•J
o
O

Q.9

2:c;S5S?5!5?35S$gSS

^^^SiftSssSSSh

P
O

z
o

oO'^NMrteO'^So©

0.S ^$$ss;sssss3

r^ d r*oooooaoaOO-^»^NN

g§822g^g§§5l$

9^^^^iOt/3U

K.S

O9<-i<-i«-4C0C0^^i4C0

^ d t*t^ooooooaao>ooo^
^^ i>xoo>ooO'-4f-ioeo

Q.S

S?§^S5SS25S2§§

tt.S

aoooftaaoo^NNM

(O (O t^ b- tt'oo 00 00 oa o o

Q.S

SSS$3$SISSSS

^ ti aoooxooo)a>oco<->'^

«0 «0 (O O t<«tt'oo 00 00 oa 9

^^g^^rs^sssss

Q.S

SSg^^gS^fSS^^

t<.b-i^i^t<«t«r«xooa)0

•OiAiOiOiOiO>O«D«DIN00

•-^d

ocooacooaoeDo»coa»o I
(O(0(0t*t*ooooooasaso

Digitized by

Google

Trblease on Design op Concrete Flat Slabs. 245

seen that the net positive moment for the central strip of the slab
is 1/40 of the gross moment, or 22,850 in.-lb. per ft.

To enable the comparison with Mr. McMillan's table, the
efifective depth of the slab will be taken as 7 in., so that the area
of reinforcement per foot width of this strip —

^«= - — J?^-^^, or 0.237 sq. in.
7 X. 86 X 16,000' ^

This will be supplied by using |-in. corrugated round bars
spaced 15 in. on centers. The negative moment for this strip is
of the same magnitude as the positive, so that it can be properly
taken care of by bending up every other bar in this strip and in
the corresponding strips of the adjacent panels.

At the center of the next strip, there is a positive moment
equal to 1/30 of the gross moment, or 30,470 in.-lb., requiring an
area of reinforcement per foot — ^il3o = 0.316 or f-in. roimds, 11 J in.
on centers.

Over the edge of the panel the negative moment for this
strip is 1/15 of the gross moment, or 60,940 in.-lb. —

An^ ^^^^ or 0.492 sq. in. p3r ft.

9X.86X 16,000' ^ ^

If every other bar from the bottom of thi?, band is bent up,
there are 0.316 sq. in. per ft. and the rema'nder, 0.172, can be
supplied in the form of short, straight bars, using f-in. rounds
spaced 21 in. on centers.

For the strip or band at the edge of the panel the net positive
moment is 1/20 of the gross moment, or 45,710 in.-lb., calling for
ilM = 0.474 sq. in. per ft., or f-in. rounds spaced 7.5 in. on centers.
At the ends of this strip over the column head the net negative
moment is 1/10 of the gross moment, or 91,415 in.-lb., requiring
-4io = 0.738 sq. in. per ft. The bent bars will supply 0.474 sq.
in. per ft., so that the remaining 0.264 sq. in. will be taken care of
by short straight bars, using f in. rounds spaced 13.5 in. on centers.

The length of the short straight bars will be such that they
will reach the fifth point of the clear span, so that they will be
10 ft. 6 in. long for this panel.

Having determined the size and spacing of the bars they can
be laid out on a drawing, or the number required estimated in
the following manner:

Digitized by

Google

246 Trelease on Design of Concrete Flat Slabs.

The balanced average of the moment factors in Fig. 22 for the
positive moments at the center of the panel is 30, so the net moment
is 30,470 m.-lb. per ft., or m the 20 ft. width of panel, 609,400
in-lb. To resist this moment will require 6.33 sq. in. of reinforce-
ment, or twenty-one f in. romid bars. The average area of the
short top bars required is 0.218 sq. in. per ft., and the total width
in which they must be supplied is 10 ft., so that 2.18 sq. in., or say
seven f-in. round bars are required.

This gives the total number of bars in one direction; in the
whole panel there are 21 straight bars 21 ft. long, 21 bent bars
31 ft. long, and 14 straight bars 10 ft. 6 in. long, all f-in. round, or a
total of 1300 lb. of reinforcement per panel. Table III gives a

Table III. — Quantity of Reinforcement Required Per Panel Under

Various Designs with a Unit Reinforcement Stress of

16,000 Lb. per Sq. In.

Method.

Thickneae
of Slab, in.

Pounds of
Reinforcement.

Ca&tQeyar

12
8
8
8
8

2,189

TiimfMiira and Mauror.

i;931

Grashof

784

Mensch

2,120

Turner.

649

McMillan

1,084

Brayton.

1.900

TrelmlM) r r . . r r

1,300

comparison of the quantity of reinforcement required under various
designs.

Fig. 23 shows the arrangement of the reinforcement in this
t3^ical panel. To avoid confusion in the drawing, the bars from
the adjoining panels are not shown. In designing this slab, it
has been assumed that the slab is fixed at the edge of the column
head, as it is the general practice to flare the head at a small angle,
making it very rigid. The head is not counted on to resist bending
moment, and is flared merely to take care of the shear and to
somewhat reduce the clear span.

To compute the deflection of the typical panel designed, the
deflection of a strip of unit width at the side of the panel will be
computed and also of a unit strip at the middle of the panel, and
the total deflection of the plate taken as the sum of these. It will

Digitized by

Google

Treleasb on Design op Concrete Flat Slabs. 247

first be assumed that these strips are freely supported, and the
deflections transformed later by the ratio of the deflection coeflS-
cients given in Fig. 12 to those for simply supported beams. The

4-1

;:i

*=:'"!:=!!;SSSSSSSE5=!!==:=-!f!

FIQ. 23. — ARRANGEMENT OF REINFORCEMENT IN TTPICAL PANEL.

computations may be very quickly made by using the graphical
solution of Mr. White's method published in Dodgers Diagrams.
In this way the deflection at the working reinforcement

Digitized by

Google

248 Trelease on Design op Concrete Flat Slabs.

stress of 16,000 lb. per sq. in. is found to be 0.15 in., or 1/1600 of
the panel side. For higher stresses the deflection would, of course,
be greater.

The unit stresses which have been employed for this panel
are very conservative, and could be safely increased to 18,000 lb.
on the reinforcement and 750 lb. on the concrete. In this particu-
lar panel the maximum concrete stress is at the colunm head and
is only 600 lb. per sq. in., while at the center of the side band
eth concrete stress is about 550 lb. With the 50-in. column head
used the shear over jd is about 100 lb. per sq. in. and as this is

MG. 24. — TEST LOAD ON PANEL IN ST. LOUIS.

pimching shear, such as in footings, a higher value could be used
if desired.

Floors designed in accordance with the methods outlined in
this paper have been built in several cities, and a few have been
thoroughly tested. In these tests, both reinforcement and concrete
stresses have been measured by extensometers. The floors tested
were designed for reinforcement stresses of 20,000 lb. per sq. in.
with the exception of the building in Minneapolis, where 18,000
lb. was used. These theoretical stresses were, of course, not
realized in the tests as tension in the concrete, arch action, etc.,
were not taken into account in the designs.

Fig. 24 shows a panel in a factory in St. Louis under a test

Digitized by

Google

Trelease on Design op Concrete Flat Slabs. 249

load of 400 lb. per sq. ft. The panel was 20 ft. i in. by 22 ft.,
designed for a safe live load of 150 lb., with an 8-in. slab. The slab
was inverted in this case, giving a perfectly flat ceiling. Two inch
caps were to be cast on the top and hidden in the cinder fill, and
the drawings showed the bars bent up into them. As actually
constructed, the reinforcement in this panel had an effective
depth over the column, the critical point of the structure, of but
60 per cent of that shown on the drawings. Even under these

FIG. 25. — ADDITION TO THE FORD MOTOR COMPANY'S FACTORY, DETROIT, MICH.

adverse conditions, the maximmn reinforcement stress over the
column head under the full test load was but 17,000 lb., and the
concrete stress at this point 950 lb. The reduction of the effective
depth over the column caused a large reduction of the moment of
inertia at that point, giving conditions approaching those in a
freely supported panel, as the stresses and deflections clearly
indicate. The deflection increased from the normal amount for a
fixed panel to 0.68 in. imder full test load, somewhat less than the
normal amount for a free span. The average reinforcement stress

Digitized by

Google

250 Trelease on Design of Concrete Flat Slabs.

at the middle of the panel was also increased in like manner to
25,000 lb. per sq. in.

Perhaps the most severe test of a floor was one made in
Minneapolis in which four panels of a warehouse floor were loaded. *
The building as laid out was not well adapted to flat slab floor
construction, because of the number and shape of the panels.
There was one row of colunms through the building, dividing the
floor into panels 13 ft. 6 in. by 19 ft. 11 in.; with the short side
resting on brick bearing walls. An 8-in. slab was used for the
200 lb. live load required. Four panels of the floor were loaded
with cement piled to prevent arching, and very complete readings
of stresses and deflections were made. Under the maximum
superimposed load of 400 lb. per sq. ft. the deflection at the middle
of the panels averaged 0.39 in., a very good figure remembering
that the panels were free at one end. The reinforcement stress
over the column head was 23,000 lb., and at the mid-point of the
span 13,500 lb. The concrete stress had a maximum value at
the column head of 1050 lb. per sq. in. These stresses show the
design to be conservative.

Fig; 25 is a view of a factory in Detroit having floors and
roof of this type of flat slab. The panels are 25 x 20 ft. and the
floors were designed for 150 lb. per sq. ft., using an 8-in. slab,
2-in. cap and 48-in. colunm head. One panel of the floor was
loaded with gravel to 300 lb. per aq. ft. Under this load, the
deflection was found to be 0.60 in. The reinforcement stress
over the column head was 13,500 lb. per sq. in. and at the center
of the slab 13,850 lb., showing a well-balanced design.

These tests seem to show quite conclusively that the method
of design outlined can be used for almost any condition of span
and loading with confidence as to the resulting strength of the
structure.

* For complete data of test see Report of Committee on Reinforoed Conorete and Buildiag
Laws. p. IOS.—Ed.

Digitized by

Google

DISCUSSION.

Mr. Alfred E. Lindau. — ^The object of the test was to gain Mr. Lindaa.
some knowledge of the action of an elastic slab, supported on
points, under a uniform load. At the time the tests were under-
taken it was not expected to obtain a great deal of information
that could be used in actual designing of concrete structures, but
rather some idea of the nature of the deformations; by plotting
the deformations obtain the deflection of the surface and perhaps
determine points of inflection, rate of change of curvature at the
various points and perhaps some idea of the relative deformation
for various points of the slab.

Many suggestions were made as to the material to be used
in making this preliminary investigation. It was suggested that
a steel plate be polished and by some mirror apparatus obtain
those deformations. Hard rubber was also suggested, plaster
plates were tried as well as a cement coated wire screen — ^all of
which gave some valuable information but did not give the
information desired and the results would not have been of any
particular interest.

Mr. Arthur N. Talbot. — ^I have been very much interested Mr. T«iboc
in the experiments made by Mr. Trelease. They show the dis-
tribution of bending moment in a homogeneous slab of rubber in
a way which could not be obtained by mere computation. The
negative moments at and between the supports, the relative
values of the positive bending moments at the center of the
panels are well brought out and there are many features of the
results which will be helpful in the application to design problems.
Of course, we have here a material that resists tension all through
it, as well as compression and one having a modulus of elasticity
much the same in tension as in compression. In the case of
ordinary flat slab construction the concrete is in compression on
one side and has a reinforcement of steel on the other to take
most of the tensile stresses. The conditions in the test must be
kept in mind because they are different from those of reinforced
concrete. Rubber is a material with a very low modulus of

(251)

Digitized by

Google

262 Discussion on Concrete Flat Slabs.

Mr.T«ibot elasticity, a material that even under small deflections would
bring in the question of catenary action, rather more than beam
or slab resistance.

I wish to ask whether in the experiments an eflfort was made
to determine the eflfect on the elasticity or on the deformations
occurring when the rubber was stretched or compressed in two
directions. This has a bearing on slab action.

Mr. Lindaa. Mr. Lindau. — There was an attempt made to obtain such

results, but some difficulty was encountered in getting satisfac-
tory compression tests on the strips used. The results on the
lateral deformation in tension were quite satisfactory and seemed
to give very uniform results, but the compressive test did not
show up quite so well. The rubber was only stretched in one
direction. The question of obtaining lateral deformation was
discussed at considerable length but tests were not carried out,
largely on account of lack of time.

Mr. Talbot. Mr. Talbot. — The matter of compressive strength is one of

considerable importance in flat slab construction, for with a
rather small percentage of reinforcement in each of four direc-
tions, or even two directions, the combined calculated stress may
nm high, unless reinforcement in compression is used, which to
my mind is not a very satisfactory arrangement with this type of
construction. It may be expected that when concrete is sub-
jected to compressive stresses in two directions, its resisting
strength will not be the same as when the pressure is applied in
only one direction. It is true — ^and this has a bearing upon the
distribution of stresses between the reinforcement and concrete
in such construction — that the amoimt of compressive deforma-
tion resulting from applying a given load is less when the load is
applied in two directions than when it is applied in one; that
seems to be the result of some experiments which we have made
and which agree with tests made elsewhere. Specimens were
made up in the form of a cross and a load applied in one direction
and also in two directions and measurements taken on the con-
crete in the cross portion and in the arms of the cross; and,
making a comparison, there seems to be less shortening in the
concrete where it is stressed in two directions, and of course a
greater expansion in the direction at right angles than when the
compressive stress is applied in only one direction.

Digitized by

Google

Discussion on Concrete Flat Slabs. 253

We have attempted to make tests of cubes by appljdng com- ^r. T«ibot.
pression against four faces instead of two. The first tests showed
considerably greater strength for loads applied in two directions
than in one, but we felt that a large part of this difference was due
to the restraint of the loading faces themselves, the friction along
the face, which kept the concrete from expanding laterally. We
then made other tests by lubricating the faces of the bearing
blocks very carefully and the values for compression in two
directions were approximately the same as those in one.

I may add that measurements were made on cantilever slabs,
slabs composed of what you may think of as the column capital
and the part of the floor slab out to the line of inflection, support-
ing them at this line of inflection, just as the load of the remain-
der of the panel would be suspended in an analysis of the flat
slab — omitting, you will see, the uniform load for the central
portion immediately surrounding the column — ^measurements of
the deformations of the reinforcement were made. We find that
in the middle line of the band the maximum stress is the greatest
at the edge of the column capital; that the stress decreases pretty
regularly from there to the edge of this cantilever slab, the line
of inflection; that through the middle portion immediately over
the column capital it decreases rapidly, although it does not
become zero with the ordinary size of bars even directly over the
center of the column; that along the edges of the bands the
maximum stress in the rods is somewhat further in, nearer the
center of the length of the reinforcing bars. The maximum
stress in the bar at the middle of the band does not differ far from
the maximum stress in the bar at the edge of the band, even
though its position differs considerably.

Digitized by

Google

THE PRACTICAL
DESIGN OF REINFORCED CONCRETE FLAT SLABS.

By Sanford E. Thompson.*

The purpose of this paper is to present material covering
the practical task of designing flat slab floors for reinforced con-
crete structures. The requisite thickness of slab, amount of
reinforcement and size of column head for different loadings and
different spans are given in Tables I-IV; and the theories and
assumptions involved in the computation are briefly discussed.
Values not included in the tables may be worked out from the
formula, finding the desired values of C^ and C, from the dia-
grams.! Curves are given also for the constants used in the
design of members with reinforcement in top and bottom, and
apply not only to flat slabs, but to any beam or slab reinforced
both in compression and tension.

For reinforced concrete buildings, the flat slab, or girderless
floor, — ^as it is sometimes called, — ^is as cheap, and frequently
cheaper, than beam and girder construction. The smooth ceil-
ings with no intersecting beams allow better distribution of the
light. The expense and complication of installing sprinkler
systems are reduced. The clear headroom for the same story
height is increased, or else, on the other hand, the story height
may be made less without reducing the effective headroom. This
last consideration alone is often important enough to dictate flat
slab floors.

With flat slab floors the entire load is supported directly on
the columns, which are usually spaced about equally in both
directions. The column heads are enlarged so as to give increased
resistance in shear and bending at the points where this is most
needed. The reinforcing bars run through the slabs over the
colunm heads in four directions, two rectangular and two diagonal.

The simplest way of considering the flat slab is to assmne

* Consulting Engineer, Newton Highlands, Mass.

t For an example of flat slab design worked out in detail see Taylor and Thompson's
''Concrete, Plain and Reinforced." 2d edition, 1011. pages 487 and 488.

(254)

Digitized by VjOOQIC

Thompson on Design of Concrete Flat Slabs. 255

that a portion of the slab extending a certain distance out from
the column is a flat, circular plate, similar to a Japanese parasol,
but with no slope to its surface. This plate is fixed to the colunm
and is assumed to extend out from it on all sides like a canti-
lever as far as the line of inflection of the slab, which line, — as
in other forms of monolithic construction, — ^is about one-fifth of*
the net span away from the support. The rest of the slab may be
considered as entirely separate from the flat circular plates but
simply supported from their outer edges or circumferences.

This is no new theory but is somewhat similar in effect to
that of a imiformly loaded, fixed or continuous beam. To illustrate
this in practical fashion, an ordinary beam uniformly loaded
and fixed at both ends will be considered. This illustration does
not in any way show the methods of determining a bending
moment in the flat slab, since, as stated below, the actual bending
moment is dependent upon the elastic tjieory. It does, however,
show quite clearly the justification of assuming the slab to be cut
through on the line of inflection.

It is known from simple mechanics that the moment at the
support of an ordinary fixed or uniformly loaded continuous beam
is Wl/12* and, at the center, is TFZ/24. Now, suppose at the
points of inflection, which also by mechanics are known to be
located at a distance 0.21 13Z from each support, the beam is cut
completely through so as to have a cantilever at each end with
a simply supported beam between. The bending moment of
the cantilever at its support, due to the load upon it, is 0.2113
TrX0.2113Z/2, and the moment at its support due to the load on
the supported beam between cantilevers, is [1—2 (0.2113)]2
WX0.2n3l. The sum of these two moments is 0.0223
m+0.0610 FZ= 0.0833 Wl or Wl/12. In other words, while
this analysis is not that which can be used for a flat slab, because
of the extra strength of the flat slab due to the multiple reinforce-
ment, the division into sections corresponds to our assumption
in the flat slab theory. In the same way it might be shown that
the center moment of the simple beam supported by the two
ordinary cantilever beams is Wl/24.

Tests of the flat slab construction at Minneapolisf indicate

* W *■ total live plus dead load. I >" distance in feet between supports.
t See paper on "A Test of a Flat Slab Floor in a Reinforced Concrete Building, " by Arthur
R. Lord, ProeeeditHf, VoL VII. page 166.

Digitized by

Google

256 Thompson on Design of Conckbtb Flat Slabs.

that the line of inflection of a flat slab floor is substantially the
same as in a fixed beam, or about ^/s the net distance between sup-
ports, although, as would be expected, the bending moment is
entirely diflferent.

Problem of Design.

The problem of the design of the flat slab, then, resolves it-
self into (1) a determination of the proper thickness and reinforce-
ment required at the support for the cantilever circular plate
supporting its own load and also the load of the rest of the slab,
and (2) a determination of the thickness and reinforcement at the
center of the span required for the simply supported section lying
between the circular plates.

Various Methods of Design of Slab.

Various methods have been advanced for the design of the
flat slab. Some are based merely on deflection tests, which give no
true basis for computations; others compute the reinforcement
carefully at the center of the slab, which is not the critical part;
others consider the construction to consist of beams between
columns with a slab between, thus obtaining ultra-conservative
results; while a plan still more common is to take the moment
at the supports arbitrarily without regard to the size of the col-
umn head. The shear or diagonal tension near the column head
is frequently disregarded altogether.

Shear at the Support.

The direct shear at the support, as in any mechanical con-
struction, is equivalent to the total load supported by the column.
This shear is readily borne by the concrete and reinforcement.
The diagonal tension, however, which, as in a beam, may be
considered as measured by direct shear, must be carefully con-
sidered. To reduce the diagonal tension and also to increase
the resistance to bending of the slab, the column head is enlarged.
To still further increase the resistance, a part of the bars in the
top of the slab over the supports may be bent down just outside
of the supports and then carried along in the bottom of the slab.

In either case, the shearing stress should be limited to definite

Digitized by

Google

Thompson on Design op Concrete Flat Slabs. 267

units, although it seems permissible to use a somewhat higher
stress than in a beam.

The diameter of the enlarged column head, which is the actual
support of the slab, should be governed by the shearing stress
either at its circumference or at a short distance outside of it.

Bending Moment at Support.

The theory of flat plates, which must be used in designing a
circular plate, is not yet clearly established. By the use of what

rr-rXL M Ada'^med iine of I f/

II 'n* — 'f^iAy mctxJmum bending i ((

\\ \\ / 7 .' / moment . \

t r^- — ^ ^f ^Theoretical t!ne \

\\ Column head^/y of inflection \
\ >v . / / ^Assumed iine v

^^>-r-j— j y<-^^ ^ of inflection ^^_

FIG. 1. — PLAN OF FLAT SLAB.

is termed, in mechanics, the elastic theory, we have a fairly good
working hypothesis. The analysis solved by Prof. H. T. Eddy*
offers, in the writer's judgment, the most rational solution of the
problem yet advanced.

In the design of the flat slab, therefore, the authorf has started
with Prof. Eddy's analysis of stresses in a homogeneous circular
plate, and from his general formulas has deduced by mathematics

* EDgineers' Society, University of Michigan, 1899.

t The author i« indebted to Mr. Edward Smuloki for the oomputationB involving intricate
analyses by higher mathematics; also to Mr. John Ayer for further studies in the practical
design.

Digitized by

Google

258 Thompson on Design of CoNCRfiTfi Flat Sla^s.

other formulas applying to circular plates free on their edges and
clamped around the columns. In a flat slab thus supported there
are horizontal stresses at right angles to each other. The eflfect of
these lateral stresses has been taken into account, this being ex-
pressed by Poisson's ratio, which is the ratio of the lateral deforma-
tion to the deformation in the direction of the stress. The value
of this ratio is taken as 0.1, which has been shown by experiments
to be a fair value for concrete of 1 : 2 : 4 proportions.

It has been found possible to reduce the complicated formulas
derived by the Eddy analysis into four formulas which are com-
paratively simple although still rather complicated for practical
use. These formulas are for four bending moments and can be

1%0 In pouftds ptraifuare foot
1 1 1 I 1 1 iiiiiil li i 1 11

^Isi^

r$etangular itett in
two tayera

q In pounds par linear foot

no. 2. — 8EcnoN of flat slab.

applied not merely to the slab at the support, but to any point in
the circular plate surrounding the column. The four moments
are as follows:

Ml = moment produced by the loading that is uniformly distrib-
uted over the circular plate and causes circumferential
fibre stress.

Jlf 2— moment produced by this same loading but which causes
radial fibre stress. .

Mo = moment produced by the loading from the rest of the slab
that is distributed along the outer edge of plate and
causes circumferential fibre stress.

Mb « moment produced by the latter loading but which causes
radial fibre stress.

Digitized by

Google

Thompson on Design of Concrete Flat Slabs. 259

A study of the analysis, however, shows that the two cir-
cumferential moments are a minimum at the support and may be
safely disregarded. The two formulas for the radial moment may
be combined and still further reduced to the following simple
form which can be used for a circle of any radius, r, within the
circular plate. The meaning of the symbols is made clearer by
reference to Figs. 1 and 2, which show the plan and the section
of a flat slab.

Let
q = uniformly distributed load around the outer edge of the

plate in lb. per ft. of length.
w « uniformly distributed load on surface of plate (including

dead load) in lb. per sq. ft.
r« = radius in feet to line of maximum bending moment (which

is within the column head).
fi = outer radius of assumed plate in feet,
r = any radius in feet where moment is to be computed; for

critical section, r is radius of column head.
Ci, Ce = constants given in Figs. 3 and 4.
Mr s total radial bending moment to be used ordinarily.
h » distance in feet between lines of inflection.

Then total radial moment at any point of plate is
Mr=wrlCi + qr^^

For convenience in computation, values of the constants

T T

C5 and C„ for various values of the ratios -^ and - are plotted

in the curves given in Figs. 3 and 4.*

With q expressed in lb. per ft. of length, w in lb. per sq. ft.,
and To in ft., the moments are in ft.-lb. per ft. or in.-lb. per in.

Position op Maximum Bending Moment and op
Maximum Stress.

As commonly constructed, the column head flares at the
top and is therefore more or less flexible. For this reason the
line of maximum bending moment will be located, not at the

* TheM Are drawn up from values in tables in Taylor and Thompson's ** Concrete, Plain
•Bd Rsiaforoed." Sd edition. 1911, pate 618.

Digitized by

Google

260 Thompson on Design op Concrete Flat Slabs.

1.0 1.1 La 13 M- 1.5 1.6 1.7 18 L9 2X)
Values of£

FIQ. 3. — ^DIAGRAM GIVING VALUES OW Cf IN VORMULA

Mr- w To* C^ + qr^ C$
foi- radius in feet to line of maximum bending moment.
n « outer radius of assumed plate in_feet.

r—any radius in feet where moment ia to be computed for actual aeotkni
ordinarily r is radius of column head.

Digitized by

Google

Thompson on Design op Concrete Flat Slabs. 261

1.0 If . L2, 13 lA L5 I.B IT L8 1.3 iO
Valuer of-|

FIG. 4. — DLIGRAII GIVING VALUES OF Cg IN FORMULA
Mr^W To* Ct + q fo C#

r«" radius in feet to line of maximum bending moment.

n » outer radius of assumed plate in feet.

r""any radius in feet where moment is to be computed for actual section ,

ordinarily r is radius of column head. ^^ j

uigiiized by VjOOQIC

262 Thompson on Design of Concrete Flat Slabs.

extreme edge of the column head, but a little within it. The
maximum stress, on the other hand, will not be on the line of
the maximum bending moment because the strength there (since
it is within the head) is increased due to the greater depth of
concrete. It is fair to assume, therefore, that the maximum
stress is at the edge of the column head, and we may assume
the "critical section" as on this line. The exact location of the
liile of maximum moment is indeterminate. Under ordinary
conditions it appears fair to assume its location as within the col-
umn head, a distance equal to the thickness of the slab. There-
fore, Mf is figured for a value of r^Vo+L In figuring this mo-
ment, values of the constants Cg and C« should be taken from the
curves in Figs. 3 and 4. As in an ordinary fixed beam, this bend-
ing moment is negative, so that the upper side of the slab is in
tension and the lower in compression. Having found the moment,
the design of the reinforcement and the thickness of the slab may
be worked out as for an ordinary beam.

The curves in Figs. 5 to 8 inclusive will be found of assistance
in working out the design.

Reinforcement in Column Head.*

The slab at the column head might be designed with the
reinforcement all in the top of the slab running in four directions
provided the slab is thick enough so that the concrete will not be
overstressed in compression. In order to reduce the thickness of
the slab and therefore save the additional cost and weight of con-
crete over the entire floor, it is economical to place reinforcement
in the bottom of the slab as well as the top, and figure it as assist-
ing the concrete to take compression. Since a portion of the
bars need to extend only far enough beyond the column head to
furnish suitable bond, the cost of this additional reinforcement
will be much less than the cost of an additional thickness of con-
crete over the entire slab. The tensile reinforcement must not
sag over the column head.

To make it easy to place the concrete and also to bring the
center of gravity of the reinforcement as near to the surfaces of
the slab as possible in order to give the longest moment arm and

* Certain features of flat slab reinforcement are covered by patents of C. A. P. Turner.

Digitized by

Google

Thompson on Design of Concrete Flat Slabs. 263

thus a thinner slab, two layers of reinforcement may be placed
in the top of the slab and two layers in the bottom. The relation
of the quantity in the top and bottom must be determined by the
design. If a thin slab is desired, even more reinforcement may
be placed in the bottom than the top. In the tables, three ratios
of reinforcement are given and the percentages selected are those
that will give the required working stresses in the concrete and
the reinforcement.

The Minneapolis test already referred to shows that not
only the remforcement directly over the column head, but the
reinforcement for a considerable distance each side takes tension.
In view of this test and of the tests made at the University of
Illinois* it is safe to assume that the reinforcement may be spaced
over a distance at least equal to the diameter of the column head
plus three times the thickness of the slab.

The determination as to whether the diagonal or rectangular
reinforcement should be placed at the top is governed by the
relative quantities of each. More reinforcement is required for
the diagonal direction through the slab, hence the layers which
are largest in section may be run diagonally.

Agreement with Minneapolis Tests.

By our theory it is possible to compute the stresses not only
next to the column head but at any point in the slab. In several
cases, knowing the exact location of the points where the deform-
ations were measured in the Minneapolis tests the stresses at
these points have been computed. Using 5.6 in. as the moment
arm, and including the radial bars as assisting to take tension,
the maximum stress in the reinforcement over the edge of the
column is 26,000 lb. per sq. in. under the normal load of 225 lb.
per sq. ft. as compared with 20,700 lb. per sq. in. given by Mr.
Lord as the actual maximum stress in the floor. This is no greater
diflference than there ought to be between design and test and
shows our method to be slightly more conservative than the
actual test.

The compression in the concrete is more difficult to check
since the exact locations of the test points are not given. Com-

♦ See paper on *' A Test of a Flat Slab Floor in a Reinforced Concrete Building." by Arthur,
R. Lord, Proceedingt, Vol. VII. page 182.

Digitized by

Google

264 Thompson on Design of Concrete Flat Slabs.

putations, however, show unquestionably that our methods are
conservative enough to allow for the irregularities in concrete
mixtures, and the danger of not having perfect concrete at the
critical section.

Moment at Center of Slab.

It is possible to adapt the Eddy theory to the design of the
center of the slab as well as to the supports. In practical design,
however, as has been indicated, the thickness of the slab is deter-
mined by the thickness at the support, which is always the greater.
But, in order to avoid too wide spacing of the bars and to adapt
the center reinforcement to that over the supports, more bars
are generally run through the slab than the results of tests would
show to be necessary. Consequently, instead of considering this
from a theoretical standpoint alone, safe values for the bending
moments may be selected, based on general principles of mechanics
and qualified by actual tests.

Let Ii« distance between lines of inflection. This distance
will be about 7 of the net span between column heads.

For the rectangular reinforcement, if the slabs between the
points of inflection were simply supported, we should have a
moment of wli/8. However, the bending moment in the Minne-
apolis tests, based on the maximum stresses under imif orm work-
ing load, is about wli/ZZ. It would appear amply safe, therefore,
to adopt a value of M^wli/\2,

For the diagonal reinforcement, the bars run in two direc-
tions, and considering both theory and test, a value of M=wli^/24t
is conservative to use for the reinforcement in each direction.

Cross Reinforcement Between Columns.

In flat slab floors cracks are apt to occur between colimms on
rectangular lines, because, since the span is shorter, the deflection
is less than in the center of the slab. To prevent these cracks, it
is advisable to place cross reinforcement of small bars in the top of
the slab.

Tables for Design of Slabs.

Tables I-IV give thicknesses of slab, reinforcement and
size of column head for various column spacings and loads.

Digitized by

Google

Design of Concrete Flat Slabs. 265

OJO ai5 020 025 030 035 040 QA5 0,50 055 060
Valued of Constant Cc
a>0.10

040 0.15 0^0 ass 0,30 0-35 040 0.4-S 0-50 a55 050
Valuer of Cons+anf Cc
a. 0.15

FIG. 5. — ^DIAGRAM GIVING VALUES OF CONSTANTS IN FORMULA
/c"^-r^ FOR a =0.10 AND 0.15

Depth of Reinforcement in Compression. Area of Reinforcement in Tension.

Q Ml pas

Depth of Reinforcement in Tension. Area of Concrete above Reinforcement.

Digitized by

Google

266

Thompson on Design of Concrete Flat Slabs.

OiO OJS OJZO a25 O^Q 035 040 045 0^0 055 a60
Vdlues of Con3+cint Cc
a*0.ao

0.04a

aiO OJS 0^0 0.^5 030 0.aS ft40 a45 050 055 0.60
Vdlues of Cons+cinf Cc
a=0.a5

FIG. 6. — DIAGRAM GIVING VALUES OP CONSTANTS IN FORMULA

/c"^,^ FOR a =0.20 AND a =0.25

Depth of Reinforcement in Compression. Area of Reinforcement in Tension.

Depth of Reinforcement in Tension. Area of Concrete above Reinforcement.

Digitized by

Google

Thompson on Design op Concrete Flat Slabs. 267

Three arrangements for reinforcement over the column head
are chosen: The first where the area of reinforcement in the top
is twice the area of reinforcement in the bottom; the second
where the two are equal; and the third where the area of rein-
forcement in the bottom is one and a half times that in the top.
This gives the designer a variety of thicknesses of slab. The

aio ais 020 0.25 oao 035 C40 045 050 055 ojbo

Votued of Cone+Gin+ C©

FIQ. 7. — DIAGRAM GIVING VALUES OF CONSTANTS IN FORMULA

/*-;

FOR a "0.30

a">

P-

Cch^

Depth of Reinforcement in Compression.
Depth of Reinforcement in Tension.
Area of Reinforcement in Tension.
Area of Concrete above Reinforcement.

percentages of reinforcement selected are those which produce,
with the given conditions, a compressive stress of 800 lb. per
sq. in. in the concrete and 16,000 lb. in the reinforcement. In
order to allow 800 lb. in the concrete, it should be mixed in pro-
portions as rich as 1 part cement to 2 parts fine aggregate to 4
parts coarse aggregate. Poisson's ratio is assumed as 0.1, which
from recent tests appears to be a fair value.

Digitized by

Google

268 Thompson on Design of Concrete Flat Slabs.

The size of column head has been figured for a shear of 60 lb.
per sq. in. on a circle a distance, t, (the thickness of slab) outside
of the column head. This shear is used simply as a measure of the
diagonal tension. The value is somewhat larger than is permitted
in beam design but appears to be warranted in the case of flat slabs.

a040r

ooos

0,04(1

Valuer of Conatan+ C^

no. 8. — ^DIAGRAM GIVING YALUBS OF C0N8TANTB IN FOBlfTTIiA

Depth of Reiaforcement in Compression.
Depth of Reinforcement in Tension.
Area of Reinforcement in Tension.

P =

Area of Concrete above Reinforcement.

The reinforcement in the center of the slabs has been figured
for a stress of 16,000 lb. per sq. in.

Diagrams for Designing Slabs.

To provide for cases not covered by the tables, curves for
values of Cg and C, are given so that the moment under various
conditions can be readily figured from the formula for the bending
moment given above.

Digitized by

Google

Thompson on Design op Concridtb Flat Slabs. 269

Diagrams for Determining Reinforcement in Top
AND Bottom of Beams or Slabs.

In Figs. 5 to 8 curves are plotted for finding the values of
the constants C^ and C, in the formulas for the reinforcement and
concrete stresses in beams or slabs with reinforcement in top
and bottom. The curves are drawn for diflferent values of a,
the ratio of depth of reinforcement in compression to depth of
reinforcement in tension, and for diflferent values of pVP> where
p = ratio of cross-section of reinforcement in tension to concrete
above it*, and p^ ^ ratio of crossHsection of reinforcement in
compression to this same area of concrete.

EXAMPLB.

Example. — ^For a warehouse floor with a live load of 150
lb. per eq. ft. and a column spacing of 20 ft. each way, what is
the necessary thickness of slab, size of column head, and amount
of reinforcement?

Sohdian. — ^From Table II the thickness of slab is given as
8} in., the size of column head as 5.5 ft., and the area of rein-
forcement as 24.7 sq. in. at top of slab and same amount at bottom
of slab over column, using ratio of area of reinforcement in ten-
sion to area of concrete below as 0.017. Dividing these values
by 4, as each end of the bands is effective, we have 24.7/4 = 6.2
sq. in. as the area of reinforcement in each band. For this may
be used twenty |-in. round bars spaced 5 in. center to center for
both tension and compression reinforcement.

The amount of reinforcement required at center of rectangular
band is 0.17 sq. in. per ft. of width. Placing a f-in. round bar
every 10 in. gives more than the necessary area, but ease in plac-
ing the reinforcement makes up for the extra amount. The
amount required at center of diagonal band is 0.35 sq. in. per
ft. of width. |-in. round bars every 10 in. will thus give neces-
sary amount.

* Where the tenaile reinforoement ia at the top, aa over a support of a flat slab or beam,
the concrete area is taken below the tensile reinf oroement.

Digitized by

Google

270 Thompson on Design op Concrete Flat Slabs.

Table I. — Design of Flat Slab.

Thickness of Slab, Areas of Reinforcement and Sizes of Column

Head are Given for Different Spans and Percentages of

Reinforcement.

LIVE LOAD 100 LB. PER SQ. FT.

ft.

it)
in.

Q
ft.

•a.9

8q. in.

aq. in.

aq. in. aq. in.

12
12
12

14
14
14

16
16
16

18
18
18

22
22
22

0.014
0.017
0.022

20 0.014
20 0.017
20 0.022

0.014
0.017
0.022

24 0.014
24 I 0.017
24 0.022

0.007
0.017
0.033

5
5i

7i
6i
5i

8i
7i
6i

8i
7i

lOi
9
8

10
8J

2.00
2.00
2.50

2.25
2.75
3.00

3.00
3.25
3.75

3.50
3.75
4.50

4.00
4.50
5.00

4.50
4.81
7.26

5.94
7.93
9.96

9.51
10.95
14.01

12.48
14.42
18.67

16.36
19.47
23.89

4.50
5.00
5.75 I 31.05

20.21
I 24.05

5.00 25.10
5.75 I 30.41
6.50 I 37.80

2.25

4.81

10.90

2.97

7.93

14.95

4.76
10.95
21.05

6.24
14.42
28.00

8.18
19.47
35.80

10.11
24.05
46.60

12.55
30.41
56.60

0.16
0.17
0.18

0.19
0.20
0.21

0.22
0.23
0.24

0.26
0.27
0.28

0.30
0.31
0.32

0.34
0.34
0.35

0.38
0.39
0.40

0.09
0.09
0.09

0.11
0.11
0.11

0.12
0.12
0.12

0.14
0.14
0.14

0.16
0.16
0.15

0.18
0.17
0.16

0.20
0.20
0.19

* Area of reinforcement over column head » circumference of column head in inchea
Xd Xp or p' depending upon whether the reinforcement is in tension or compression. This
reinforcement is assumed as distributed over the entire widths of the bands. Thus if a band
of reinforcement has 2 sq. in. in section the area, efTective, for two bands will be 4 aq. in. (See
example.)

Digitized by

Google

Thompson on Design op Concrete Flat Slabs. 271

Table II. — Design op Flat Slabs.

Thickness of Slab, Areas of Reinforcement and Sizes of Column

Head are Given for Different Spans and Percentages of

Reinforcement.

LIVE LOAD 180 LB. PER SQ. FT.

12
12
12

14
14
14

16
16
16

18
18
18

20
20
20

22
22
22

24
24t

d a ^

■ffi

(p)

0.014
0.017
0.022

0.014
0.016

(pO

0.007
0.017
0.033

6.007
0.017
0.033

007
017
033

007
017

0.033

0.007
0.013

iff

ID.

7i
6i
5i

8i 10

<
(0

8}
7}

lOi

n 11*

8i lOit

ft.

2.25
2.50
3.00

S o

aq. in.

5.64
6.81
8.72

3.00 , 8.72

3.50 I 11.22

3.75 I 13.22

3.50 I 12.03

3.75 13.83

4.50 I 17.75

4.00 I 15.32

4.50 18.05

5.50 ' 24.00

5.00 21.80

5.50 ! 24.70

6.25 I 31.14

5.50 I 26.15

6.25 I 31.07

7.00 39.24

7.00 36.05

7.00 36.95

I-

aq. in.

2.82 ' 0.18

6.81 0.19

13.09 , 0.21

4.36 0.22

11.22 0.23

19.82 0.24

6.02 ' 0.26

13.83 0.27
26.80 0.28

7.66
18.05
36.00

10.90
24.70
46.70

13.08
31.07
58.80

18.03
29.57

0.31
0.32
0.33

0.34
0.35
0.36

aq. m.

0.10
0.10
0.10

0.12
0.11
0.11

0.14
0.13
0.13

0.16
0.15
0.17

0.17
0.17
0.16

0.38 I 0.18
0.39 0.17
0.40 0.17

0.42
0.43

0.21
0.20

The values printed in black type are figured for a column head 7 ft. in diameter and the.
thickncM of the alab is increased to withstand the shear.

* Area of reinforcement over column head — circumference of column head in inchea Xd X P
or pi dei>ending upon whether the reinforcement is in tension or compression. This reinforce-
ment is assumed as distributed ovei the entire widths of the bands. Thus if a band of reinforce-
ment has 2 sq. in. in section the area, effective, for two bands will be 4 sq. in. (See example.)

t The thickness of slab for the 24-ft. span may be decreased to 8} in. by using 0.022 and
0.033 ratios of reinforcement and bending the bars to resist diagonal tension.

Digitized by

Google

272 Thompson on Design op Concrete Flat Slabs.

Table III. — Design of Flat Slabs.

Thickness of Slab, Areas of Reinforcement and Sizes of Coltthn

Head are Given for Different Spans and Percentages of

Reinforcement.

LIVE LOAD 200 LB. PER 8Q. FT.

GQ
ft.

.- P 2 «
.23 !£

(P)

i
i

(lO

(0
in.

ft.

2 2

0) Q.

£a

sq. in.

ill

la I

sq. in.

aq. in.

12
12
12

14
14
14

16
16
16

18
18
18

20
20
20

22
22 1

24t

0.014
0.017
0.02!^

0.014
0.017
0.022

0.014
O.017
0.022

0.014
0.017
0.022

0.014
0.017
0.019

0.014
0.016

0.014

0.007
0.017
0.033

0.007
0.017
0.029

0.007
0.012

0.006

6
6
4i

6
4i

9i
8i

lOi

6i
5i
5

7J
6i
6i

8i
7i
6

8
7

lOi
8i
7i

lot

12t

2.50
3.25
3.76

3.25
3.75
4.50

4.00
4.50
5.50

4.75
5.50
6.50

5.50
6.25
7.00

6.25
7.00

7.00

6.60

9.38

10.37

10.31
12.02
15.88

14.26
16.60
21.70

18.80
22.92
29.70

24.71
29.08
31.35

31.38
34.90

37.90

3.30

9.38

15.40

5.16
12.02
23.80

7.30
16.60
32.55

9.46
22.92
44.60

12.36
29.08
47.85

15.69
26.15

16.24

0.20
0.21
0.23

0.24
0.25
0.26

0.28
0.30
0.32

0.33
0.34
0.35

0.37
0.38
0.39

0.42
0.42

0.46

0.10
0.10
0.10

0.13
0.12
0.11

0.14
0.14
0.14

0.17
0.16
0.15

0.18
0.17
0.16

0.20
0.19

0.22

The values printed in black type are figured for a column head 7 ft. in diameter and the
thickness of the slab increased to withstand the shear.

* Area of reinforcement over column head— circumference of column head in inches
Xd Xp or p' depending upon whether the reinforcement is in tension or compression. This
reinforcement is assumed as distributed over the entire widths of the bands. Thus if a band
of reinforcement has 2 sq. in. in section the area, effective, for two bands will be 4 sq. in. (Seo
example.)

t The thickness of slabs for the 22- and 24-ft. spana may be decreased to BH u^* i^d 9H
in. respectively, by using 0.022 and 0.033 ratios of reinforcement and bending the bars to i
diagonal tension.

Digitized by

Google

I^OMPSON ON Design of Concrete Plat Slabs. 273

Table IV. — ^Design of Flat Slabs.

Thickness of Slab, Areas of Reinforcement and Sizes of Column

Head are Given for Different Spans and Percentages of

Reinforcement.

LIVE LOAD 300 LB. PER SQ. FT.

•s

)n of re-
ompres-
! below
tension.

1
n head. I

rcement
tension.

int over
ireasion.

IP
III

lag-:
his

2*- 1

2 -^ a

!•§:§

(d)

°.2

•

He!

ft.

(p)

(p')

in.

ft.

sq. in.

sq. n.

0.014

0.007

3.50

9.72

4.86

0.24

0.12

0.017

4.25

12.27

0.25

0.11

0.022

0.033

5.00

15.56

23.30

0.26

0.10

0.014

0.007

4.25

14.03

7.02

0.29

0.13

0.017

5.00

16.84

0.30

0.12

0.022

0.033

6.00

21.18

31.72

0.31

0.11

0.014

0.007

5.00

18.48

9.24

0.33

0.15

0.017

6.00

23.10

0.35

0.15

0.022

0.033

7.00

29.05

43.50

0.35

0.12

0.014

0.007

6.00

24.59

12.30

0.39

0.17

0.017

7.00

29.18

0.40

0.16

0.015

7.00

26.74

0.39

0.15

0.014

0.007

10}

7.00

32.32

16.16

0.43

0.19

20t

0.015

0.011

lOit

7.00

33.65

25.36

0.43

0.18

22t

0.012

0.003

10}

12it

7.00

32.61

8.23

0.47

0.22

24t

0.010

0.000

13i

ISit

7.00

35.64

0.00

0.51

0.25

The values printed in black type are figured for a column head 7 ft. in diameter and the
thickness of the slab increased to withstand the shear.

•Area of reinforcement over column head = circumference of column head in inches
X<f Xp or p' depending upon whether the reinforcement is in tension or compression. This
reinforcement is assumed as distributed over the entire widths of the bands. Thus if a band
of reinforcement has 2 sq. in. in section the area, effective, for two bands will be 4 sq. in. (See
example.)

fThe thickness of slabs for the 20-. 22- and 24-ft. spans may be decreased to 8>i in.. 9H
in. and 11 in. respectively, by using 0.022 and 0.033 ratios of reinforcement and bending the
bars to resist diagonal tension.

Digitized by

Google

DISCUSSION

Mr. Lindau. Mr. Alfred E. Lindau. — Regarding this method of design-

ing a flat slab or plate of concrete it might be of value to note
that the analysis is based on the assumption that there exists
a line of contra-flexure around the colmnn substantially circular
in form and near the quarter point of span, that the balance of
the span is supported along this line and that consequently there
is no stress in any direction along this line. There is some ques-
tion as to this line of contra-flexure, in fact there is some evidence
to show that instead of a line of contra-flexure there exist per-
haps only four points where there is no stress in any direction.
If such is the case the analysis would be entirely diflFerent from
that which has been outlined.

Mr. Andenon. Mr. W. P. Anderson. — One of the important things in
designing a flat slab has not been touched — ^the bending moment
on the exterior column. I think that has an effect on the thick-
ness of the slab. The thicker the slab, it appears to me, the less
that bending moment would be. I have not gone into the matter
thoroughly enough to determine how the bending moment of
the exterior column is to be figured, but it seems to me that it is
the critical point in the design of the flat slab and one that has
not yet been touched upon enough by those who have given the
subject thought. This point in the design of the flat slab has
not been covered in the tests made on the flat slabs. They have
been made in the center of the building and the exterior colimm
has not been tested at all to see what the moment is. It is some-
thing that ought to be covered. I am not prepared to say how
it ought to be taken care of, but I believe it ought to be looked
into.

Mr. Lindau. Mr. Lindau. — The test covered in Part II of the report of

the Committee on Reinforced Concrete refers to an end or wall
panel. In fact, it was a building with one line of colimms making
all panels end panels, so to speak.

Mr. Green. Mr. HERBERT P. Green. — I have been experimenting a

little bit this last year on flat slabs of a different construction

(274)

Digitized by

Google

Discussion on Concrete Plat Slabs. 275

from any others I have seen. Realizing that the shear around Mr. Green,
the head of the column is the greatest thing to be considered, I
endeavored to work out a way of thickening the slab at this point
and have succeeded in obtaining a flat ceiling without the flar-
ing head. We have built one building by this system and tested
it and it is a perfect success.

The reinforcing bars run directly and diagonally from col-
umn to column. The bending moment on each beam is considered
as coming from the load of one-fourth of the panel, which posi-
tively occurs because triangles are formed in between the direct
and diagonal girders which are filled with ordinary stock tile.
Between these tile are concrete joists, supported by the beams.
This construction can be figured by any engineer without em-
pirical formulas.

At the center of the slab between the columns on the diagonal
girders in the bottom or top of the slab, supplementary rein-
forcement can be placed, but it is generally not necessary. Every
beam and every girder, or every joist and every beam, as they
may be called, is of a T-section, the concrete above the tile form-
ing the flange of the direct and diagonal beams and the supple-
mentary joists in the triangles, so that the analyses of the stresses
in this slab are very easy, according to the ordinary theories of
design.

Mr. Lindau. — I would like to ask Mr. Green whether in Mr.Lindau.
his method of flat slab construction he has made joints along
these triangles, so that the tile filler is disconnected absolutely
from the balance of the slab thereby making separate beams,
or whether it is all concreted in and becomes a portion of the
slab as a whole.

Mr. Green. — ^The tile actually becomes a portion of the Mr. Green,
slab, but — according to the methods generally used for figuring
the efficiency of tile joist construction — the tile is considered as
nothing but a core, so that the concrete construction is the prin-
cipal thing considered. If we should remove the tile, in this
construction it would be the same thing as removing it in any
tile joist construction, L e., we would have a floor with ribs or
stiffeners below, and in this case they would be in four directions
from column to column, with supplementary stiffeners in the
triangles. The tile is simply a form.

Digitized by

Google

276 Discussion on Concrete Flat Slabs.

Mr. xjadaii. Mr. Lindau. — I take exception to the assumption that

you have a series of beams that are free and independent to act
as beams; because just as soon as you connect up the slab so as
to have anything resembling a homogeneous structure you have
changed the stress condition from a slab to a beam. The loads
are transmitted by bending moment or transverse stresses, or a
combination of those two; you cannot draw imaginary diagonals
or anything on that slab and say that this is the direction in
which the load will travel. The load will travel along the lines
of relative rigidity of the structure and you cannot make it go
in one direction rather than another unless the slab is built accord-
ingly. This is the difficulty with most theories and methods of
strip analysis, where the freedom of the slab is tied up in various
ways and consequently cannot be separated into the various
elements it may be considered to be made up of.

Mr. orMn. Mr. Green. — ^That may be so, but in the two-way reinforced

concrete floor with tile fillers one figures the loads are carried
through the joists, which are supported on the beams, to the
beams which run directly between the columns. Now I simply
make two more beams or headers in the slab. I have run them
diagonally across from column to column and the joists in the
triangles between the direct and diagonal beams are supported
by these direct and diagonal beams. It is the same principle
that would occur in a wooden framed structure with diagonal
beams, except that we have a T-section in all directions and
the stresses in the concrete above and below the neutral axis
at any point can be figured just as readily as in the two-way
reinforced concrete floor.

Digitized by

Google

THE DESIGN OF CONCRETE GRAIN ELEVATORS.

By E. Lee Heidenreich.*

During the International Congress of Engineers in 1893, at
Chicago, III., I was appointed by the American Society of Civil
Engineers to present a paper on '* American Grain Elevators,"
having at that time had about ten years' experience in the con-
struction and erection of grain elevators in North and South
America. Elevators at that time had the same functions as they
have today and may as a rule be divided into certain classes:

1. Farmers' elevators or small station elevators with a
capacity from 5000 to 50,000 bushels — ^where the farmers would
deliver their grain in wagon loads, up to 5 tons capacity, have
it weighed, elevated and shipped out in railroad cars.

2. Mill elevators or cleaning elevators, built as adjuncts
to flour mills, within varying capacities from 50,000 to 500,000
bushels and containing, besides receiving and shipping apparatus,
scales and cleaning machinery.

3. Terminal elevators, divided into storage and working
houses. The working houses would receive the grain, weigh
and ship it into the storage elevators or cars. The storage
elevators would merely be for the purpose of storing grain.

4. Transfer elevators located at prominent points where
different grains are graded and changed from one grade to
another and shipped from western to eastern cars.

5. Marine elevators, where the grain would either be received
from canaJ boats or vessels or shipped into them. These
elevators are either dock elevators or floating elevators.

Up to 1893 there were no reinforced concrete grain elevators
in the United States. At that time two large grain elevators
were imder construction at Galatz and Braila, Roumania, and
Mr. Herman 0. Schlawe, a representative of the Roumanian
government, sent over to see how we handled grain and took care
of oiu" grain elevators, showed me the plans of the same elevators
designed by Luther of Braimschweig, Germany. These elevators

^Consulting Engineer. Kansas City, Missouri.

(277)

Digitized by

Google

278 Heidbnreich on Concrete Grain Elevators.

were of unit construction, somewhat similar to the one described
by Mr. Darnell,* excepting that the bins were honeycombed
hexagonal cells, built of flat slabs 3 ft. square, and comer pieces,
all interlocked. They were all built on the ground, erected and
interlocked somewhat similar to tile construction. The difficulty
with flat slab walls was, of course, that they were subject to
direct flexion, and therefore became quite cumbersome, owing to
the lateral, rather heavy pressure in grain bins, and were only
suitable for bins of small dimensions.

Steel tanks have been tried in this country, but there was
more or less sweating of the tank, which to some extent damaged
the grain. Circular reinforced concrete tanks were then designed,
using the space between the tanks whereby considerable economy
of ground area was effected as compared with isolated tanks and
thereby increasing their combined strength and carrying capacity
on the soil or the sub-structure. The introduction of the new
design was, however, exceedingly difficult.

The owners thought the tanks would sweat and damage the
grain and builders thought the tanks would surely burst. From
1896 to 1899 I designed some scores of cluster tank elevators and
presented them to millers and grain elevator owners entirely in
vain, imtil in 1899, Frank H. Peavey, of Minneapolis, as an
experiment built a tank 23 ft. in diameter 130 ft. high, filled it
with grain and found that it held grain without spoiling. He then
sent some representatives over to Europe to see what effect
grain storage in reinforced concrete had in Europe; they examined
the tanks and grain elevators at Galatz and Braila, in Roumania,
and came back with a report which resulted in the construction,
in 1901, of a large one million and a half bushel elevator in
Duluth. In the meanwhile I had built, in 1900, a cement storage
elevator at South Chicago, consisting of four tanks 25 ft. in
diameter and 56 ft. high, clustered in such manner as to utilize
the space between them.

It must be remembered that wheat weighs 50 lb. per cu. ft.
while cement weighs 100 lb.; and the fact that a cement tank with
5-in. walls at the top and 7-in. at the bottom, 25 ft. in diameter,
would hold cement, went a long step towards convincing people
that it was safe to store grain in circular tanks and also in cluster

* See p. 4G4.— Ed.

Digitized by

Google

Heidenr£ich on Concrete Grain Elevators. 279

tanks. The tanks in South Chicago were erected on top of
reinforced concrete girders, which in turn rested on columns,
thus forming a working house underneath where the bagging of
the cement took place.

Since that time cluster tanks have grown up like mushrooms
all over this country, Canada, South America and Europe.
There are upwards of 60,000,000 bushels capacity built in the
United States alone to-day and they are being built in every
part of the country for terminal elevators and largely for milling
elevators, giving safe and good storage and eliminating the ques-
tion of insurance and the calamity due to fire or burning up of
the storage plant.

The design of reinforced concrete grain elevators, consisting
of circular tanks, has been carefully studied. It is of course
entirely a function of the. grain pressure in rest and in motion.
Experiments and calculations have been made by a number of
well-known engineers. Janssen, Wilfred Airy, Tolz, Prante,
Jameson of Canada and Milo B. Ketchimi, have all made experi-
ments and developed formulas whereby the grain pressure may
be determined at the different heights of the bin. Grain is not
like a liquid — ^Professor Ketchum calls it a semi-liquid, — the
pressure line forms a curve.

Inasmuch as Janssens' solution tallies very closely with
Jameson's experiments, we niay write the lateral pressure

and the vertical pressure

Cm'

^(■-" * )

Where ty= the weight per cu. ft. of grain or 50 lb. for wheat,
jB=bin area divided by its perimeter (hydraulic radius),
ft' = the coefficient of friction of grain against bin surface;
A=the height of the bin, and

C=the ratio between the lateral and the vertical pressure;
e = being the base of the Naperian logarithm or 2.71828.

According to Jameson for wheat C=0.6 and for grain on
concrete ft' =0.4 to 0.425.

Digitized by

Google

280

Heidenbeich on Concrete Grain Elevators.

For quick and easy calculation the lateral pressure per sq.
ft.^qwh, where q is the ratio of grain pressure to liquid pressure
and

F--^ -1.667 gwA
Fig. 1 shows the value of q for different ratios of height
WHEAT PRESSURES IN GRAIN BINS-

\23^5ST6B\0

PIG. 1. — VALUES OP Q POB DIPPBRENT RATIOS
OP HEIGHT TO DIAMETER OP BIN.

divided by the diameter or width ^-r. The maximum bottom

h
pressure occurs when t =3.6.

Fig. 2 shows a series of circular bins with intervening spaces.
The lateral pressure per sq. ft. of the grain in a circular bin at a
depth h is equal to

L^qwh

Digitized by

Google

Heidbnbbich on Concrete Grain Elevators.

281

and the tension in the reinforcement reqxiired for 1 ft. in height is

The reinforcement area in sq. ins.
where /« is the working stress of the reinforcement per sq. in.

WIQ. 2.-HSEBIBB OF CIRCULAR BINS.

If, however, -the circular bin is empty and the two opposite
interstices A and B filled, we have a condition as shown in Fig. 3
where it is quite apparent that instead of single reinforcement of
the bin, there should be reinforcement both at the intrado and
extrado of the ring.

Compression in the direction A B clearly causes tension at

Digitized by

Google

282

Heidenreich on Concrete Grain Elevators.

the extrado at C and Z>, and at the intrado A and B, From
experiments in loading culvert pipe firmly supported at the two
lower quarter points corresponding to E and F, the maximum
moment may approximately be written (Fig. 4) :

.s^^t-

FIG. 3. — PLAN OF CIRCULAR BIN.

The usual formula for the resisting moment is

M«/C6d«andX«iM(l-|)=p/,(l.|)

where

/, and /c the unit stresses in reinforcement and in concrete.

Digitized by

Google

Heidenreich on Concrete Grain Elevators.

283

The stress in the concrete or in the reinforcement may be
expressed by

-5'('-l)

bd*

■('-!)

bd*

By adding 1 in. to di or (is, the ring thickness is found.

Example: The 25 ft. diameter bins are 80 ft. deep and sur-
romided by interstices, find the dimensions of concrete wall and
reinforcement required at 60 ft. from the top.

h 25

H A— -

hence
Then

and

no. 4.

9-0.23 and L=0.23X60X60=6901b. per sq. ft.

T.m^^^^,^,

8625
"^•"Tfinno ** '^^ sq. in. per each foot high

The inner and outer reinforcement from the eflfects of grain
in the center bin is

-~ =0.27 sq. in. or say §-in. rounds 8§ in. on centers
Vertical rods J-in. rounds 24 in. on centers tied to horizontals.

Digitized by

Google

284 Heidenbeich on Concrete Grain Elevators.

For external pressure we have

^ - 59 -6 hence g "0.1 andL^O.lX 50X 60-300 lb. per eq. ft.
10

(^ -o) L'D' C-?^, -6) 300X 26 X 12

jif . W2 L = Vl._41_/ 18700 in. lbs.

64 64

^ ^ / 18700' / 18700 __^.
*"*'\ibXl2 "\ 110X12" ^•^^'"•

hence the wall should be 5 m. thick.

18700
'^•*' 0r8fiy^l6o6oS^4 ""^-^^ ^^- ^^' ^^^ ^^^» ^^^^ extrado and intrado,

showing that the stresses from the filled interstices are greater
than those resulting from the filled circular bins. Some con-
tractors using single reinforcement in the center bins have rodded
the interstices, others increase the thickness of the bin walls and
the connection between contiguous circular bins.

The logical method, would, however, seem to be to use double
reinforcement, a practice the author invariably prefers for circular
culverts.

As to the calculation of square bins, this is a simple matter,
after the lateral pressure L has been found.

The bending moment will be in in.-its.

LP
Jl/ « — 12, where Z= width of bin in feet.

The resisting moment as before

■V"

M^UxJP and 4= ^,,,

Area of reinforcement required

A« = 17-; as before

Interior bin walls having alternately pressure on either side
are, of course, reinforced on both sides. The connection of the
reinforcement at the intersections and the comers of the bin walls
become of the greatest importance. Where possible the author
prefers to employ a strong wire fabric as a part of the reinforce-

Digitized by

Google

Heidbnreich on Concrete Grain Elevators.

285

ment for the outer walls, nmning this reinforcement horizontally
atoiind the entire building.

Fig. 5 shows a common country elevator of 15,000 bushels
capacity. The structure rests upon a slab or raft extending
under the entire building and connected with the side walls so as
to make the pit absolutely waterproof. The concrete is mixed

nte

30-0-

\ ,

P""^

FIG. 6. — CONCRETE COUNTRY ELEVATOR.

1 : 2 : 4 to maximimi density and to each bag of cement is added
5 lb. of petroleum residuum oil which has been found an excel-
lent and cheap method of making the walls and roof impervious
to moistiu^e. The roof and cupola walls are reinforced with
fabric to prevent cracks by shrinkage.

Digitized by

Google

286

Heidenreich on Concrete Grain Elevators.

Fig. 6 shows an outside wall connection and the reinforce-
ment, illustrating how the horizontal and hooked members are
tied to the vertical rods, to insure their proper position until

FIG. 6. — OUTSIDE WALL CONNECTION OF REINFORCEMENT.

the concrete has been poured. The hooks of the comer rods
(Fig. 7) are placed by springing the vertical rods until the comer
rods can be hooked in and thereby form a support for the outside
carrying rods.

FIG. 7. — METHOD OF HOOKING CORNER REINFORCEMENT.

In 1911 the com crop of the United States was. . 2,531,488,000 bu.
and of wheat 631,388,000 "

Total of 3,152,786,000 "

Digitized by

Google

Heidenreich on Concrete Grain Elevators. 287

Most of this crop was handled through farmers' elevators,
averaging, say 150 cars a year, or 150,000 bu., meaning a total
of approximately 20,000 farmers' grain elevators in the country.
Most of these are built in wood and will ultimately have to be
replaced with concrete construction, which in turn shows one
immediate channel for the use of Portland cement. The cost
of a 150,000-bushel grain elevator in concrete is about $5;000.00
exclusive of machinery and millwrighting. Considering the
thousands of wooden elevators rotting away and burning up along
our western trunk and grain lines — ^the almost irreparable loss
to a town or community of farmers when its elevator bums down,
the ajonual cost of insurance and maintenance of these wooden
elevators — ^the great doctrine of conservation brings into the
limelight the impending want of reinforced concrete grain
elevators in every nook and comer of the land.

Digitized by

Google

DISCUSSION;

Mr.Lindau. Mr. Alfred Lindau. — In the design of circular bins for

exterior pressure of the space between the bins, is account taken
of the support that is afforded by the adjacent bins? As I under-
stand, the worst condition is where the two opposite spaces would
be filled with grain and others would not be filled with grain;
but the bin, I understand, is supported on the other diameter by
the adjacent bins. Is this taken into consideration in the strength
of the structure, or merely left as an added factor of safety;
What are the general principles that govern the economy of grain
storage? Bins are built high and you can build them wide,
circular or square. Has an investigation been made to show the
design or adopt a bin that would attain economy for storage
pmposes?

Mr. Heidenreich. Mr. E. Lee Heidenreich. — ^Thc location of the adjacent
bins is taken into consideration, otherwise we could use Profes-
sor Talbot's formula divided by 16, vxl/lQ, in place of dividing by
64 — ^an empirical factor that I brought in.

There are many items, many functions, which come into
play in the determination of a grain elevator. First, the storage
capacity required. It is possible to store grain in very large
bulk, requiring very large bins. I have built them up to 30,000
and 40,000 bushel bins for winter storage. In other cases there
are what we call pocket bins, carload bins, which are required
where certain shippers want their grain individually, a practice
quite common. Then again comes the floor space. In a city the
ground space is very expensive and it is important to economLee,
and elevators are built as high as possible, as high as the soil will
carry them. If the soil will not carry them they are put on
timber piling under all piers. In other places, again, you can
spread the bins considerably, but, as a general rule, to utilize the
gravity of the grain in handling it throughout the elevator is
considered the best economy. For instance, in a cleaning elevator
you build it so high that from the top of the elevator head the
grain runs first into the gamer, then into the scales, then into the

(288)

Digitized by

Google

Discussion on Concrete Grain Elevators. 289

cleaning bin; from there through the cleaners and separators to Mr. Heidenreich.

receiving bins and then into the loading apparatus. But you gain

by it, because the grain goes by the gravity from the top in one

process to the bottom in the place of being re-elevated, which

consumes power and of course adds to the cost of handling of the

grain.

There are a great many other elements entering into shape of
grain elevators, but these are a few of the functions that must be
considered.

Mr. Wm. M. Kinnet. — In bms holding cement, there is Mr. Kinney
often a sweeping action, similar to a wave action. Mr. Heidenreich
spoke of the semi-fluid condition of grain, or an action like a
semi-fluid. In a cement bin the cement will hang up on one side
and all at once give way and sweep clear to the other side, across
to the opposite side wall of the tank, and put a strain there that
would have to be figured on. I was wondering whether there was
anything like that in the bins.

Mr. HEmENREiCH. — This only happens with wet oats; they Mr. Heidenreich.
act very much like cement in bulk. On account of that action
instead of building the hoppers converging to the center in a
cement storage elevator at South Chicago, they were built diverg-
ing to the sides, like over a peak in the center of the bin, whereby
the hoppers diverge downward toward the circumference, so that
the cement in sliding down the hopper would, as it were, spread
itself around to the edges. That seemed to help considerably
against the wedging proclivity of the cement at the top of the
hopper.

Mr. F. L. Williamson. — In connection with the storage Mr. wiuiamson.
of cement it might be interesting to note that at our plant, cement
is being stored in circular tanks, in interface tanks and also in
square tanks, all of reinforced concrete construction. The storage
tanks are 80 ft. deep and 30 ft. in diameter. The interspace
tanks, of course, are of corresponding size and three square tanks
are 8 ft. square and 64 ft. deep. They have been in use for two
years and have proven satisfactory. The strain Mr. Kinney
speaks of has no doubt been exerted many times and with no
damaging effects.

Digitized by

Google

REPORT OF THE COMMITTEE ON MEASURING
CONCRETE.

Engineers and contractors for reinforced concrete structures
have long felt the need of some standard method of measuring
quantities in contracts for concrete work. This is especially
the case in drawing up and adjusting unit price contracts. At
present, although mo^t engineers and contractors measure and
estimate on the same general principles, they nearly all differ
in the details of their work, with the result that disputes occur
and sometimes serious loss is occasioned or injustice done. A
feeling exists that some of the principles or fundamental rules
of measurement are founded on a wrong basis and need careful
consideration and revision. With this object in view this Com-
mittee was appointed last year to consider the matter and now
submits its report.

In submitting these Proposed Standard Methods the Com-
mittee wishes to call attention to the purpose for which they are
intended and the principles by which they were guided in fram-
ing them.

The purpose is, first, to establish a correct method for award-
ing unit price contracts and for measuring up work performed
under the same; second, to inform concrete contractors and
engineers of the best methods of estimating their work and work-
ing up unit costs; third, it the above two objects are attained
there is likely to be greater uniformity in published cost data.

The principles which have guided the Committee in draw-
ing up these rules are of great importance and are as follows:

First, — All work shall' be measured net as fixed or placed
in the building or structure, and therefore material cut to waste,
voids, temporary work, etc., shall not be allowed for in measure-
ment but in price.

Second. — In no case shall non-existent material be measured
to pay for extra labor in different parts, but such diflScult or
expensive parts or extra labor shall be separately measured
and described.

(290)

Digitized by

Google

Report of Committee on Measuring Concrete. 291

Third, — That all the chief items of labor and material entering
into the cost of concrete work shall be separately measured and
described by units that correctly represent the labor and material
involved.

It follows from the first and second of these principles that all
work shall be measured as it stands, and that forms or extra labor
in placing concrete work shall not be allowed for by doubling or
trebling the measurements of concrete, and that laps in reinforce-
ment should be measured.

The acceptance of the third principle involves the recognition
of the several items entering into the cost of a mass of concrete
as separate and distinct operations, viz:

(1) Concrete mixing and placing.

(2) Forms.

(3) Reinforcement.

(4) Surface finish.

The separation of forms from concrete is not in opposition to
the first principle of omitting incidental work, as forms should be
considered as an item of labor. The labor of supporting wet
concrete by means of forms is entirely distinct from the labor
of mixing and placing concrete and is done by different men at a
higher rate of pay. It is a distinct operation in the progress of
the work and not a labor incidental only to the placing of the
concrete.

The committee realizes that the proposed methods if generally
adopted will be used and administered by inspectors, superin-
tendents and foremen, as well as by engineers, architects and
contractors. There are some who will search for opportunities
for taking imfair advantages of the methods for their own profit
while many will catch the spirit and use them fairly. It is, there-
fore, very necessary that care should be taken to provide no oppor-
tunity for fraud or unfair dealing.

In framing methods of measurement the Committee has felt
that it is not necessary to give instructions as to how to measure,
but only as to what units should be used in measuring and what
items should be included or left out of such measurements. For
instance, it is necessary to say whether I-beams shall be deducted
or not deducted from the mass of concrete work, but it is not

Digitized by

Google

292 Report of Committee on Measuring Concrete.

necessary to give a method as to how to measure the irregular thick-
ness of a tunnel lining or of the backing to a masonry wall. The
latter are matters of mensuration and outside the scope of this
report.

The Committee has endeavored to keep to methods of measure-
ment only as distinct from methods of cost, and in general no
instructions are given as to the way to fix the price or cost of any
item, or as to how to make up any prices for work measured under
these methods. For instance, the cost of plant is added by some
contractors to the per yard cost of concrete while by others it
is taken as a lump sum at the end of their estimate Nothing
is said about this and in the committee's judgment should not be.

The Committee wishes to lay very special emphasis upon the
classification of concrete, forms, reinforcement and surface finish
into separate items. It is realized that especially in the case of
forms a radical departure from the present method of measuring
imits is recommended, but after careful consideration the Committee
feels convinced that the present method of including forms and
finish and sometimes steel in the cubic measurement of mono-
lithic concrete is fundamentally wrong and should be altered.
A few instances will be cited to make this point clear.

Supposing a contractor takes a contract for a building with
unit prices per cubic foot for concrete in floor, columns and foun-
dations, the said unit to include the cost of forms and granolithic.
If the engineer decides to reduce the thickness of the floor slab
from 5 to 4 in. the contractor has to put in just as much form work,
but is paid for only 80 per cent of the form work he estimated
upon. If the building was increased from two stories to three
stories in height and floors were finished in granolithic he would
have to finish twice as many floors with granolithic as he estimated.
Or vice versa, suppose a wall shown 8 in. thick is increased to 12 in.
thick, the owner has to pay for 50 per cent more forms and surface
finish than before, although the amount done is the same. In a
dispute recently settled in the courts a contract provided for a
certain price per cubic yard being paid for concrete foundations,
this price covering the cost of concrete, forms and steel. In the
execution of the work it was found necessary to make the footings
deeper than shown, but most of the steel was omitted. The
contractor claimed that though the steel was omitted his contract

Digitized by

Google

Report op Committee on Measubing Concrete. 293

price per yard of concrete was the one he was entitled to be paid
on even if some of the steel was left out. The court upheld his
view and he received judgment accordingly.

These instances are t3rpical of the gambling nature of the pres-
ent system of letting and taking contracts for concrete at an
inclusive unit price per cubic foot, and the Committee feels con-
fident that if the methods suggested are adopted they will be
of real value to the community, eliminating such uncertainties
and inequalities. Very few contractors in the present day like
to take these gambling contracts.

In other respects the Conunittee has endeavored to conform as
far as possible to the general practice in this country and desires
to make it standard.

Some of the detailed points involved in the methods submitted
will now be considered.

I. Monolithic Concrete.
(a) Concrete.

The rule stating that concrete in diflferent parts of the build-
ing or structures shall be measured and described according to
its accessibility and location and the purpose of the work would
indicate that there should be separate measurements appearing
in the bill of quantities for:

Concrete in footings.

Concrete in columns.

Concrete in floors, beams and girders.

Concrete in paving.

Concrete in basement walls.

Concrete in curtain walls and partitions.
And so on according to the nature of the work.

Concrete in mass foundations.

Concrete in abutments.

Concrete in arch ribs.

Concrete in spandrel walls.

Concrete in bridge floors and beams.

Concrete in parapet walls and cornices.
And so on according to the nature of the work.

Digitized by

Google

294 Repobt op Committee on Measuring Concrete.

In the event of any of these items, such as columns, foot-
ings, etc., being of different mix on different floors or places,
these also would be measured separately (see Appendix I).

The matter of placing bolts, inserts, pipes, etc., in the con-
crete was considered. It is not covered by any rule, as the Com-
mittee held that it was not strictly a concrete item. The placing
of such items should be taken by the number or the lineal foot
as the case may be and not allowed for in the price of concrete
or forms. The cost of placing is part of the cost of the item
itself, and as such would come under the heading of plumbing,
steel work and so on.

(6) Forms.

After careful consideration the Committee feels that the
square foot of the surface supported is the correct unit for the
measurement of forms. In their judgment the item which should
be measured and paid for is the operation of supporting the wet
concrete imtil it is set. This item is practically an item of labor
although material enters into the cost of same.

It is the practice of some firms to estimate forms by the
amount of lumber used, but this is not a correct unit of measure,
not only on the theoretical ground stated above, but on the practi-
cal ground that no two contractors would use the same amoimt
of lumber in erecting a piece of form work, and if lumber on imit
price contracts were measured and paid for by the board foot
there would be an incentive to a contractor to use more lumber
than was necessary to do the work.

It is necessary to give a definite unit which everyone knows
how to measure. It is not possible to determine beforehand
just how much lumber will be used in any piece of form work.
It is not possible to accurately measure the amount of lumber
used during the work, and it is not possible to take this amount
from the lumber bills because all lumber bought on a job does
not go into forms, some of it going into temporary buildings,
sheeting and shoring and many other necessary parts. For
instance, when the sheet piling of trenches is done, the lumber
often goes into forms, and of 1,000 board feet of roofers delivered,
half may go to the forms and half to temporary buildings. Fur-
ther, much lumber is used two and three times. Other parts

Digitized by

Google

Report of Committee on Measuring Concrete. 295

are used but once only and some pieces are sawed off as waste
and not used at all, and if a rule were to be framed that would
measure lumber, how should all these difficulties be provided for.

Paying for lumber on a unit price contract gives to an un-
scrupulous man an opportunity to claim a higher payment from
the owner, or in the case of cost data, for a foreman to deceive
his employers as to the cost of the form work he is doing. A
superintendent can put in twice as much bracing and posts as
he actually needs and make his costs look low, and if the con-
tractor is paid by the board foot he also would receive payment
for unnecessary work.

In contrast to these difficulties the square foot measure-
ment of form work does not alter and is easily and quickly meas-
ured; and further, even those who estimate by the board foot
have to determine the number of square feet to be supported
before computing the number of board feet to be required in
the building.

In the measurement of iSoor forms the sides of beams are
added to the measurement of under side of slab and beams,
although some contractors take a flat measurement of floor sur-
face. The latter method would work out unjustly in the case
of a change being made in the depth of the beams, such change
working to the detriment of either the owner or the contractor,
while if the method that the Committee recommend is adopted
such a change will adjust itself.

Some discussion may be raised on the omission of any allow-,
ance for angle fillets to columns and girders, but the Committee
as practical concrete men realize that such items are a very small
part of the cost of forms and it is not a usual practice on the part
of contractors to estimate them separately. They believe that
to measure angle fillet by the lineal foot, as has been suggested,
would prove a possible source of misunderstanding in the carry-
ing out of a contract.

The separation of forms to floors, columns, footings, etc.,
follows the lines laid down for the separation of concrete items.

In the case of forms to the concrete walls poured as backing
to granite or other facing, the correct interpretation of the rules
would be to measure forms to one face only, the stone facing
doing the work of forms on the other side.

Digitized by

Google

296 Repobt or Committeb on Measttbing Concbete.

Some may question the correctness of making an allowance
for laps or passings in steel rods or fabric for reinforcement. The
Committee after careful consideration felt that these ought to
be allowed. Laps are for two purposes, one as in beam steel to
take up negative moment, and the other to provide suflScient
bond to take up the tensile strength in two bars where one bar
cannot be obtained long enough for the purpose. In the first
case no question would arise as to measuring the full length of
each rod; in the second, we think that as the specification usually
specifies the lap to be a certain number of diameters, usually
forty or fifty, there would be little difficulty on this head.

There is, of course, an opportunity for unfair dealing by a
contractor in putting in steel in short lengths. A rule not allow-
ing lap would sometimes work hardship the other way if an in-
spector insisted that steel should go in in short lengths instead
of long, and then only pay for a rod the net length of the build-
ing. An example would be in a floor of a building laid out in 10 ft.
bays where rods could be put in one length over one, two or three
bays as desired.

The CJommittee feels that the rule they have formulated
should stand. It is in accordance with the first fundamental
principle laid down and all work should be measured net as fixed
in place. They believe that any engineer would use reasonable
discretion in refusing to measure laps if it was apparent that too
many had been made for an unfair purpose.

III. Structural Cast Concrete.

The use of structural cast concrete is growing rapidly, is of
an entirely different nature than monolithic concrete, and the
Committee has endeavored to treat this part of the subject in
a way similar to which structural steel is measured and estimated.

It is, therefore, not suggested that forms be measured sepa-
rately, but recommended that erection should be separate from
making and that the unit of weight be the correct unit for measur-
ing erection rather than the unit of volume.

In measuring or computing erection quantities, it is recom-
mended that an arbitrary weight of 150 lbs. per cu. ft. be adopted

Digitized by

Google

RePOBT op COMMITTEk ON MEASURING CONCRETE. 297

(similar to the rule for steel). This method will save the ex-
pense of weighing each piece and a good deal of dispute. This
rule would not apply to cinder concrete, for which another weight
should be agreed upon.

In buildings built partly of cast concrete and partly of mono-
lithic concrete, monolithic concrete would be measured under the
rules laid down for same and the cast concrete under the structural
cast concrete rules.

The Committee doesnot recommend that grouting be measured.
This is looked upon as a small labor item incidental to the erec-
tion of the concrete members and similar in character to riveting
on structural steel work or mortar in laying cut stone.

IV. Cast Concrete Trim and Ornamental Work.

Cast concrete trim and ornamental work is more nearly
akin to cut stone work than to any other trade. The Committee
has followed the custom of the cut stone trade and based the rules
on stone mason rules, viz: to measure the smallest rectangular
solid out of which a piece can be taken for any piece of trim, in-
stead of measuring the net volume of the finished block.

Surfacing is generally done in the mold and therefore should
not be separated.

Temperature reinforcement is a very small part of the cost
and for that reason is not separated.

Trim serving any structural purpose such as lintels with
reinforcement in same to take tensile strength should be classi-
fied as structural cast concrete and measured accordingly.

As the stone mason's unit of erection is the same as the unit
by which he supplies the stone, the Committee does not suggest
that erection be separated from making.

The Committee has not at present drafted rules covering
plastering, waterproofing, concrete blocks, concrete piles, etc.,
except in so far as they are covered by the general rules governing
surface finish, concrete trim and structural cast concrete.

Plastering is understood to be any surface coating of cement
or lime mixed with fine aggregates on soffits or vertical surfaces
which is put on by hand without the use of forms.

Digitized by

Google

298 Report of Committee on Measuring Concrete.

The rules of structural cast concrete would apply to con-
crete piles cast and driven, except the driving is usually measured
by the lineal foot and not by the pound, and the Committee does
not suggest that this practice should be altered.

Respectfully submitted by the Committee on Measuring
Concrete.

Robert A. Cummings, Chairman.

L. H. Allen,

Chas. Derleth, Jr.,

H. H. Fox,

Thos. M. Vinton.

Digitized by

Google

Repobt of Committee on Measuring Concrete. 299

Appendix I.

BILL OF QUANTITIES FOR A BUILDING.

Measured by Proposed Methods.

Concrete 1 : 2| : 5.

CJoncrete 1:2:4.

Concrete 1
Cinder

Concrete 1
Forms.

1J:3.

;3:6.

Footings 600 cu. yds

Basement walls: 220 ''

Base to paving 90 "

Floors and beams 2,100 "

Colunms 650 "

Partitions and curtain walls 200 ''

WallSi roof and pent house 30 "

Cornice and parapet 100 "

Columns 110 "

Reinforcement.

FiU between screeds (2 in. thick) 225

Footings l.OOOsq.

Floor slabs and beams 15,500

C(dumns 3,500

Curtain walls and partitions 3,350

Pent house 900

Basement walls 3,000

Parapet and back of cornice 1,200

Face of cornice 36 in. girth 400 lin.

ft.

Plain round bars, in-
cluding cutting, wir-
ing, placing

li in. dia.
1 "

Plain round bars, in-
cluding cutting, wir-
ing, placing, but also
including fabricating
Square twisted bars,
including cutting,
wiring and placing.

1 in. thick, laid integral with paving

1 in. thick, laid on floors after concrete

has set 15,000

On cornice and parapet 600

On window sills 1,200

Picked face to concrete smiace 2,000

Rubbed face and cement wash one coat on concrete surface. . . 3,000

Granolithic Finish.

i
11

1x1

ix}

5,000 lbs,

7,000 •'

12,000 "

24,500 "

8,000 "

17,800 "

28,500 "

6,000 "

3,000 "

8,000 "

4,000 "

5,000 sq.

ft.

Digitized by

Google

300 Report op Committee on Measuring Concrete*

Appendix II.

BILL OF QUANTITIES FOR A BRIDGE.
Measured by Proposed Methods.

Concrete 1:3:6. Cyclopean masonry with 30 per cent

rock in foundation 2,000 cu. yds

Abutments 1,500 "

Piers 1,000 "

Abutments (upper part) 500 "

Piers (upper part) 500 "

Arch ribs 1,200 "

Columns above same 150 "

Spandrel walls and wing walls 250 ''

Bridge floor and beams 600 *'

Forms Abutments (below finished grade) 12,000 sq. ft.

Abutments (above finished grade) 6,000 ''

Piers (below finished grade) 10,000 "

Piers (above finished grade) 8,000 "

Arch ribs 15,500 '*

Columns 1,200 "

Spandrel walls and wing walls 5,000 ''

Bridge floor and beams 12,000 "

Parapet wall 2,500 "

Coping of same 300 lin. ft.

Reinforcement. Plain round bars 1 Jin 30,000 lbs.

" " " 1 " 50,000 "

" i'' 75,500 "

" " " 1" 12,000 "

" " " I" 10,000 "

Crandalled Finish. Piers and parapet 2,500 sq. ft.

Rubbed Finish. Arch ribs (face and soffits) 40,000 "

Granolithic Finish. 1 in. thick to sidewalks to bridge floor

laid integral with slab 2,000 "

Curb and gutter of same 300 lin. ft.

Sidewalk. To approaches with 4 in. base 1 : 21 : 5
and 1 in. top and includes cinder foun-
dation 800 •'

Curb and gutter to last, curb 10 in. high,
gutter 12 in. wide including forms and

finish done in one operation 200 "

Extra for rounded corners to same 3 ft.

girth 4

Cast Concrete. Balusters 8 in. x 8 in. x 2 ft. high, includ-
ing forms and steel and setting in place 50
Coping to same 18 in. x 9 in. with molded
edges, including forms, steel and finish
and setting in place 200 lin. ft.

Digitized by

Google

PROPOSED STANDARD METHODS FOR THE
MEASUREMENT OF CONCRETE WORK.*

The following divisions are recognized as separate and dis-
tinct items in the construction of concrete work for which separate
modes of measurement are necessary.

I. Monolithic Concrete:

(a) Concrete.
(6) Forms.

(d) Surface Finish.

II. Sidewalks.

III. Structural Cast Concrete:

(a) Concrete.

(6) Reinforcement.

IV. Cast Concrete Trim and Ornamental Work.

The following general rules shall govern the measurement
of the above items (with the exceptions where specifically
noted) :

(a) All work shall be measured net as fixed or placed in the
structure.

(6) In no case shall non-existent material be measured to
cover extra labor.

I. Monolithic Concrete.
(a) Concrete.

1. The unit of measure for all concrete shall be the cubic foot.

2. In no case shall the measurement of concrete be held to
include the forms.

*'Tbe proposed method* ■ubmitted by the Committee were disouwed. refened baok to
the Committaa aod appear here u amended. — Ed.

(801)

Digitized by

Google

302 Standard Methods for Measuring Concrete Work.

3. AH concrete shall be measured net as placed or poured in
the structure.

4. In no case shall an excess measurement of concrete be
taken to cover the cost of forms or extra labor in placing.

5. All openings and voids in concrete shall be deducted with
the following exceptions:

(a) No deduction shall be made for reinforcement, I-beams,
bolts, etc., embedded in concrete except where a unit has a sec-
tional area of more than 1 sq. ft.

(6) No deduction shall be made for pipes or holes in concrete
having a sectional area of less than 1 sq. ft.

(c) No deduction shall be made for chamfered, beveled or
splayed angles to columns, beams and other work, except where
such chamfer, bevel or splay is more than 4 in. wide measured
across the diagonal surface.

6. Each class of concrete having a different proportion of
cement, sand or aggregate shall be measured and described
separately.

7. Concrete in the different members of a structure shall be
measured and described separately according to the accessibility,
location or purpose of the work.

8. Concrete with large stones and rocks embedded in same
(cyclopean masonry) shall be measured as one item and described
according to the richness of the mix and the percentage of rock in
same.

9. Concrete in stairs shall be measured by the cubic foot
and shall include surface finish when the mixture is the same
throughout.

(6) Forms.

10. The unit of measure for form work shall be the square
foot of actual area of the surface of the concrete in contact with
the forms or false work.

11,' Forms shall in every case be measured and described
as a separate item and in no case shall the measurement of concrete
be taken to include forms.

12. No deduction shall be made in measurement of surface
of concrete supported by forms, because of forms being taken
down and re-used two or three times in the course of construction.

Digitized by

Google

Standard Methods for Measuring Concrete Work. 303

13. The unit price for superficial measurement of forms
shall be deemed to include the cost of struts, posts, bracing,
bolts, wire ties, oiling, cleaning, and repairing forms.

14. No distinction shall be made between wood and metal
forms.

15. Forms to different parts of a structure shall be measured
and described separately according to the position in the structure,
accessibility, purpose and character of the work involved.

16. No allowance shall be made for angle fillets or bevels
to beams, columns, etc., but curved moldings shall be measured
and described separately as hereinafter provided.

17. No deduction in measurement of forms shall be made
for openings having an area of less than 25 sq. ft.

18. No deduction shall be made in floor forms for heads
of columns of any shape.

19. No deduction shall be made in column and girder forms
for ends of girders, cross beams, etc.

20. No allowance shall be made for hand-holes in column
forms for clearing out rubbish.

21. The measurement of column forms shall be the girth
of the four sides or circumference multiplied by the height from
the floor surface to the under side of floor slab above,

22. Forms to octagonal, hexagonal and circular columns shall
be measured and described separately from forms to square col-
umns.

23. Caps and bases to columns and other ornamental work
shall be measured by number and fully described by overall
dimensions.

24. The measurement of beam forms shall be the net length
between columns multiplied by the sum of the breadth and twice
the depth below the slab, except for beams at edge of floor or
around openings which shall have the thickness of floor added
to the sum of the breadth and twice the depth.

25. Wall forms shall be measured for both sides of concrete
wall.

26. Allowance shall be made by number for pockets left
for future beams.

27. Moldings in form work shall be measured by the lineal
foot.

Digitized by

Google

304 Standard Methods for Measuring Concrete Work.

28. Forms to circulax work shall always be measured sepa-
rately from forms to straight work.

29. No measurement or allowance shall be made for con-
struction joints in slabs, beams or arch ribs, to stop the day's
concreting.

30. Construction joints or expansion joints to dams and
other large masses of concrete shall be measured by the square
foot as they occur.

31. Forms to cornices shall be measured by the lineal foot
and the girth stated. (The term girth shall be taken to mean
the total width of all curved and straight surfaces touched by
the forms.) Plain forms to back of cornice to be measured
separately.

32. Forms to window sills, copings and similar work shall
be measured by the lineal foot.

33. Forms to the upper side of sloping slabs such as saw
tooth roofs shall be measured whenever the slope of such slab
with the horizontal exceeds an angle of 25 degrees.

34. Forms to the under side of stairs shall be measured by
the superficial foot.

(a) Forms to the front edge of the stairs shall be measured
by the lineal foot.

(6) Forms to the ends of steps shall be measured by number.

35. The unit of measure of reinforcement shall be the weight
in pounds.

36. The weight shall be calculated on the basis of a square
rod 1 in. X 1 in. x 12 in., weighing 3.4 lb.

37. Steel rods for reinforcement shall be measured as the
net weight placed in the building.

38. Deformed bars shall be measured separately from plain.

39. No allowance shall be made for rolling margin.

40. No allowance shall be made for cutting or waste.

41. No allowance shall be made for wire ties, spacers, etc.

42. No separation shall be made according to accessibility,
location and purpose of reinforcement except in special cases.

43. In measuring reinforcement the rods shall be measiured
by the lineal foot as laid. All laps shall be allowed for.

Digitized by

Google

Standard Methods for Measuring Concrete Work. 305

44. The rods of each dififerent size shall be measured and
described separately.

45. Bent bars shall be measured separately from straight
bars.

46. Pipe sleeves, tumbuckles, clamps, threaded ends, nuts
and other forms of mechanical bond shall be measured separately
by number and size and allowed for in addition.

47. Wire cloth, expanded metal and other steel fabrics
sold in sheets or rolls shall be measured and described by the
square foot. The size of mesh and weight per square foot of steel
in tension shall be stated. No allowance shall be made for waste,
cutting, etc., but all laps shall be measured and allowed for.

(d) Surface Finish.

48. The unit of measure for finish of concrete surfaces shall
be the square foot. Finish shall always be measured and described
separately.

49. No measurement or allowance shall be made for going
over concrete work after removal of forms and patching up voids
and stone pockets, removing fins, etc.

60. Granolithic finish shall be measured by the square foot
and shall include all labor and materials for the thickness speci-
fied.

51. Finish laid integral with the slab shall be measured
separately from finish laid after the slab has set.

52. No allowance shall be made for protection of finish with
sawdust, sand or tenting.

53. Grooved surfaces, gutters, curbing, etc., shall be meas-
ured separately from plain granolithic and shall be measured by
the square foot or lineal foot as the case may require.

54. The following shall be measured by the square foot:

Cement wash. (State how many coats.)

Rubbing with carborundum.

Scrubbing with wire brushes.

Tooling.

Picking.

Plastering.

Etc.

Digitized by

Google

306 Standard Methods for Mbasumng Concrete Work.

II. Sidewalks and Pavements.

55. Sidewalks and pavements shall be measured by the
square foot.

56. The one measurement shall include concrete, finish, lin-
ing in squares and cinder or stone foundations.

57. Curbs and curb and gutter work shall be measiu'ed by
the lineal foot and separated according to character and size,
and shall include foundations, forms, finish and cost of special
tools if any.

58. In measuring curbs the full height and width or thick-
ness of same shall be taken, but the measurement of sidewalks
shall also be taken the extreme width of horizontal surface.

69. Circular comers to curbs and gutters shall be measured
separately by number, stating radius and length measured on
the curve.

60. Vault lights shall be measured by the square foot, the
measurement to include glass, forms, steel and finish. Beams
under vault lights shall be measured by the lineal foot. In measur-
ing vault lights the measurement shall go at least 4 in. beyond
the outside line of the glass in each direction.

III. Structural Cast Concrete.
(a) Concrete.

61. The term structural cast concrete is taken to mclude
unit construction by the various systems.

62. The unit of measurement for structural cast concrete
shall be the cubic foot, and shall be measured net as provided
for monolithic concrete.

63. The various members shall be measured on the ground
before erection.

64. No measurement shall be taken of forms.

(6) Reinforcement.

65. Reinforcement shall be measured separately as pro-
vided in Paragraphs 35 to 47, inclusive.

66. The unit of measure for the erection of structural con*
Crete shall be the weight of the finished member in pounds.

Digitized by

Google

Standard Methods for Measuring Concrete Work. 307

67. In measuring the erection of structural cast concrete
having a crushed stone or gravel aggregate, the concrete shall
be assumed to weigh 150 lb. per cu. ft.

68. No measurement shall be taken of the grouting in
structiu-al cast concrete. It shall be deemed to be covered in
the price of erection.

IV. Cast Concrete Trim and Ornamental Work.

69. Cast concrete trim shall be measured by the cubic foot,
but the measurement shall be the smallest rectangular solid
that will contain the piece measured and not its actual content.

70. No allowance shall be made for forms.

71. No allowance shall be made for reinforcement in trim
and ornamental work.

72. No allowance shall be made for surface finish in trim
and ornamental work.

73. Circular work shall be measured separately from other
work.

74. Mitre blocks and end blocks for cornices, etc., shall
be measured separately from straight molded work.

75. Vases, seats, pedestals, balusters and other similar
items shall be measured by number and description with over-
all dimensions.

Digitized by

Google

CONCRETE RETAINING WALLS.

By J. M. Meade.*

Concrete walls are now mostly built of two types, viz., plain
and reinforced. They are very popular for track elevation, depres-
sion and dike work, especially where ground is valuable and they
replace earth dikes to economize space, etc.

PLAIN MONOLITHIC CONCRETE RETAINING WALLS.

The plain concrete walls are designed for what is known
as gravity section, being heavy enough so that their weight and
stability will stop them from overturning. The best authorities
figure the width of base of such walls as 0.45 up to 0.65 of their
height, varying the width according to circumstances. Where a
retaining wall of this type is built up close to the end of ties, as
on an elevated road, it becomes a surcharged wall. When designed
as 0.45 1 have seen them fail by pushing over and would recommend
not less than 0.65 of the height for a surcharged wall, finishing
18 in. wide at the top. The common practice in railroad work
of using arbitrary ratios of width of base to height of walls tends
to cause a neglect of the study of the proper distribution of the
pressure on the foundation and it seems to be difficult to get away
from such practice. It is a well-known fact that movement from
the original alignment, due to unequal settlement, is the most
common cause of failures or defects. The writer has in mind some
flagrant cases of this kind in. the City of Chicago that have caused
:he owners a very heavy expense. This question is one of great
importance and each particular case should be carefully investi-
gated and studied, so the amount and distribution of the pressure
on the foundation may be accurately determined.

Many walls of poor design have come to the attention of the
writer, there being entirely overlooked, due to a lack of analysis
of the design, the most effective section and minimum amount
of material for an economic design.

* Engineer, EaBtern Linee, Atchison, Topeka and Santa F6 Railway, Topeka, Kan.

(308)

Digitized by

Google

Meade on Concrete Retaining Walls.

309

In constructing retaining walls, it is of the greatest importance
that serious thought be given to the matter of earth filling and
embankments behind the walls. The drainage is quite easily
accomplished by filling or placing close up to the back of the wall
some open or porous material, such as crushed or refuse stone,
large size gravel, brick bats, etc. ; cinders will also do considerable
good. Weep holes should be placed in the wall of 3 or 4 in. drain

FIG. 1. — REINFORCED CONCRETE TYPE OF RETAINING WALL AS USED IN RAIL-
ROAD WORK.

tile, vitrified, about 15 or 20 ft. apart, according to conditions,
extending the blind drains to a point near the top of the wall if
circumstances seem to warrant. It is also quite important in
plain concrete retaining walls, to use expansion joints about 30 ft.
apart, of the dovetail pattern. If this is not done temperature
stresses will crack the wall. Such expansion joints are a good
investment.

Digitized by

Google

310 Meade on Concrete Retaining Walls.

reinforced concrete retaining walls.

Reinforced concrete retaining walls consist of either a thin
vertical plane attached to a horizontal base and well braced by
counterforts on the back and buttresses in front or they may
be designed as cantilevers, in which case the wall is connected to
a wide base resembling an inverted T.

It has been found in actual practice that reinforced concrete
walls are more economical than the plain monolithic gravity walls,
as the material in the latter type cannot be fully utilized for the
reason that the section must be made heavy enough so that the
dead weight and size prevent overtmning. On the other hand,
in reinforced concrete walls a part of the retained material is
used to prevent overturning and the wall only need be made strong
enough to withstand the moments and shears due to the earth
pressure. The wall is lighter and exerts less pressure on the ground,
which, with the opportunity of extending the base of the wall,
often enables the builder to use ordinary foundations instead of
piles.

Reinforced walls allow the use of a more scientific design
than the gravity walls and have been known to be more reliable
than the plain concrete. It is quite common practice to make the
base of these reinforced or cantilever type of walls about 0.60 of
their height and then reduce in size about as shown by Fig. 1.

Digitized by

Google

DISCUSSION.

Mr. Willis Whited. — ^I would like to ask Mr. Meade if ui. wwted,
the long dike at Topeka is built open, exposed to the action of
floating ice and, so far as he can judge by the experience so far
had, whether it is necessary to thicken the wall to resist the impact
of floating ice. We have some work in Pennsylvania on some of
our highways where it is necessary to use a very long dike which
will be exposed almost to the top in flood time and there would
be a good deal of impact from floating ice against it.

Mb. John M. Mbadb. — ^In a great many instances the Mr. Meade,
reinforced concrete wall has proven very economical in cities.
At Topeka, where the floods of 1903 were so disastrous, the
river was diked on one side with an earth dike and on the other
side with a reinforced concrete wall, which was largely planned
on account of the valuable right of way. It is one of the longest
concrete dikes in this country, being about a mile in length.
The wall was finished late last fall and of course we have not had
any tests yet. It is 18 in. thick at the top. The river has only
been up to above normal or high water a few feet, and the wall is
set back well from the bank, in some places 30 or 40 ft., so it
would take a big flood to get above the banks and reach this wall.
With a heavily reinforced wall the ice would shear off so rapidly,
going parallel with the wall, as to cause little pounding.

This work was especially watched with reference to its cost
of construction, etc., being a new departure in dikes. Another
departure was the use of Joplin chatts. There have been some
misgivings as to the reliability, for heavy work, of chatts, a by-
product in the manufacture of zinc. The chatts can be had for
about the cost of loading, which is about 15 cents a cu. yd. at
Joplin and of course railroads do not figure the freight. I had
occasion to look at a pier built over the Chikaskia River in the
southwestern part of Kansas on a line used jointly by the Santa F^
and the Frisco Raiboads. A stone pier had failed about eight
years ago and was re-built with Joplin chatts. There was a dam
below the bridge, so that the pier has been in water from 6 to 8 ft.
deep all the time. I was very favorably impressed with the results
of the use of chatts at that place.

(311)

Digitized by

Google

REINFORCED CONCRETE PILES.
By Robert A. Cummings.*

It is well established that concrete piles are a satisfactory
substitute for wooden piles in increasing the stability of founda-
tions, whether they are of the cast-and-driven type or the casi-
in-place type. The cast-and-driven type is made of reinforced
concrete, molded to the desired shape and cured before being
driven. The caat-dn-place type is made by forming a hole in
the ground and filling it with concrete.

The history of the cdstrand-driven type is intimately con-
nected with the development of reinforced concrete and in the
same manner has been embarrassed by patent litigation. In
March, 1907, a final decision was reached in the British High
Courts of Justice that the fundamental idea of the reinforced
concrete pile was covered by Brannon's patent of 1871 and that
subsequent improvements must be limited to the details of
reinforcing.

At the Chicago Convention of this Association in 1909,
the writer described certain tests and methods of reinforcing
for increasing the unit value of concrete in compression. It is
the application of one of these methods and other practical
improvements that form the subject of this paj)er.

It is nearly ten years since the writer began the design
and manufacture of concrete piles. During this period, by a
process of elimination, an efficient method of reinforcing has
been developed, which is correct in design and economical in
cost.

METHOD OF REINFORCEMENT.

All piles of the cast-and-driven type must be reinforced with
longitudinal rods, because the pile is hoisted and handled by a
line from the pile driver which is fastened at or near the butt.
Consequently, the pile must sustain its own dead weight while
being raised, as well as shocks and impact against obstacles,

* Consulting Engineer, Pittsburgh, Pa.

(312)

Digitized by

Google

CuMMiNos ON Reinforced Concrete Piles. 313

before reaching its position in the leads of the pile-driving machine.
.With the point on the ground and the other end being elevated
the pile must act and be designed as a beam supporting its own
dead weight and shocks. The limiting proportions of depth or
thickness of pile to the unsupported beam length produces heavy
tensile and compressive stresses, with considerable deflection.

It has been observed that when handling a pile of this type,
it rarely fails in compression in the concrete, but cracks are
usually discovered on the tension side. These cracks can be
accounted for by the slipping of the usual longitudinal rods used
in the concrete while the pile is being hoisted. While such cracks
are not suflSciently serious to condemn the pile, they may affect
the permanency of the reinforcement.

In order to overcome this defect all longitudinal rods should
be anchored at the ends, those at the butt opposite each other
being bent over into a loop and welded together, while those at
the point are all brought together and electrically welded into
one piece. Twisted and deformed rods are . advantageous for
longitudinal reinforcing, as the allowable bond stress is higher
than for plain rods.

The uniform circumferential spacing of longitudinal rods is
very important, because any side of the pile may be subjected
to tensile stresses and whichever side is in tension there must
be sufficient reinforcement in position to take the strain. The
circumferential spacing of the longitudinal rods can be secured
by means of a special spacing device placed at intervals of about
5 ft. throughout the length of the pile.

The hooping of concrete adds greatly to its ability to resist
axial loads. Therefore, longitudinal reinforcing should have a
helical wrapping of wire throughout the length of the pile, the
pitch of which must not exceed 3 in. This wire wrapping will
assist in taking care of diagonal stresses resulting from the
handling of the pile.

Practical experience indicates that the butt end of the pile
which receives the impact of the hammer should be especially
reinforced. This has been done by means of a special reinforce-
ment consisting of a unit cage of flat bands, held 2 in. on centers
by a spacing bar for a distance of 2 ft. In the plane of each
band a flat wire spiral is fastened to the cage. The embedment

Digitized by

Google

314

CxTifMiNGs ON Reinforced Concrete Piles.

of this unit cage in the butt of the pile forms a resilient cushion
to receive the impact of the hammer. In no case, in the driving
of thousands of piles with this cushion, has the butt itself been
broken.

The general design of the pile is shown in Fig. 1.

METHODS OF DRIVING.

Piles of the aistrandrdriven type are handled and driven by
means of an ordinary pile driver. The ability to vary the fall
of the drop hammer is of great utility in overcoming the variable
resistances to be met with in the driving. It has been found

r-^*~-^

snoops I'x^'

XSoffSkelSpirai
mlpper4Ringi

Na 5 Spirvl mrt S 'hfch
FIG. 1. — ^DESIGN OF REINFORCED CONCRETE PILE.

advisable to increase the weight of the ordinary drop hammer
in ratio of weight of hammer to weight of pile of from 2 or 3
to 1, so that the weight of the drop hammer for driving the
concrete piles will vary from 7000 to 12000 lb.

Steam hammers are not as efficient or desirable for driving
concrete piles as are drop hammers. This was shown last fall
on a contract when a test was made between a steam hammer
and a heavy drop hammer, under the same conditions, using
the same kind and size of concrete pile. The heavy drop ham-
mer did not break a pile, whereas the steam hammer broke sev-
eral below the cushion. Further, the steam hammer did not
drive as many piles in a given time as did the drop hammer.

Digitized by

Google

CuMMiNGs ON Reinforced Concrete Piles. 315

The following explanation is offered as it applies also to
the driving of the heavy steel casings or core used in the making
of the cast-dn-place type of pile: The limited fisdl (3 ft.) and light
ram (3000 lb.) of the steam hammer^ while 'delivering twice as
many blows per minute as the drop hammer, loses a large part
of its energy in overcoming the proportionately heavier weight
of the pile, steel casing or core.

The following analytical treatment by Mr. Barton H. Coffey,
of New York, confirms and justifies the practice of the writer
in using hammers as heavy as 12,000 lb. for driving concrete
piles and it may be asserted with some feeling of confidence that
if penetration is to be the gauge for measuring the supporting
power of a pile, the ratio of the weight of the hammer to the
weight of the pile or core that is driven must be taken into con-
sideration.

The advantage of using a heavy hammer is evident from the
following analysis:

Let M * Mass of hammer=—

9
" TT = Weight of hammer.

" X = Weight or mass of pile in per cent of hammer.

" y «8 Velocity of hammer on striking pile.

" Vi = Common velocity of hammer and pile.

Momentum of hammer = ilf 7 (on striking pile.

" " " and pile at common velocity = Af (i-fX) Vi.

These are equal, provided no external force acts, which we
will assimie for the present is the case.

Then ilf F = M (1+X) Vi] therefore Fi = fx v ^^^

The energy in hammer on striking pile is

»JJIfF» (6)

The energy in both hammer and pile at common velocity is
}ilf (1 +X) V* which upon substituting (a) becomes } JIf r-r^ (c)

The difference between (6) and (c) represents the loss of
kinetic energy in the system at point of common velocity or
greatest compression. In other words, the percentage of the

Digitized by

Google

316 CuMMiNGs ON Reinforced Concrete Piles.

original energy of hammer that has gone into compressing pile
and hammer^ distorting or crushing them and into heat, or
broadly, internal work done upon each other by the striking
bodies. Obviously, the smaller this percentage of internal work
the less liability there is of crushing or distorting the pile. To
resume

(c)-(6)- } ilf V« - i JIf ^ - J ilf F« (l-jij^) W

This equals the internal work.

The following table gives internal work for various values
of X in per cent of the total kinetic energy of the hammer.

X Internal Work.

0.25 20 per cent

0.50 33

0.75 43

1.00 50

1.50..... 60

2.00 67

3.00 75

If the hammer weighs 4 tons and the pile 1 ton, there will
be 20 per cent internal work at maximum compression, whereas
if the pile weighs 8 tons the internal work will be 67 per cent.
There is an external force acting against the pile, t.e., the fric-
tional and displacement resistances of the earth.

Two extreme cases may be assumed limiting all others.

1. There is no external force. There the internal work
will be simply that necessary to overcome the inertia of the pile
and put it in motion. In this case the table is rigorously accurate.

2. The external force is great enough to prevent any move-
ment of the pile. In this case the entire kinetic energy of the
hammer goes into the internal work and the relative weights of
hammer and pile are immaterial.

All intermediate cases where movement occurs are a com-
bination of (0 and (2), where obviously it is advantageous to
employ a heavy hammer.

Digitized by

Google

CuMMiNGS ON Reinforced Concrete Piles. 317

MANUFACTX7RE.

The procedure in constructing cast-andrdriven piles com-
mences with the preparation of the molding bed. This, of course,
will vary with the site, but it is desirable to select a flat and
convenient location near the place of driving. It is very important
that the bed shall be stable, so that settlement due to the weight
of the piles is avoided. Where the ground is soft and yielding,
pine stakes 2 in. x 4 in. x 3 ft. long, pointed at the ends, are
driven to a solid bearing. These stakes are located at intervals
of about 4 ft. in each direction and the tops cut to a uniform
level. Then, 4 x 4 in. pine sills are toe-nailed to the top of the
stakes in longitudinal rows about 4 ft. on centers. Upon these
sills a 2-in. solid wooden floor is placed, which forms the molding
bed. It is desirable that this bed shall be uniformly level to
receive the forms for the piles.

The forms are made of two pieces of 2 x 8-in. dressed pine,
battened together and placed on edge to form the sides of the
pile. The bevels or angles for tapered or octagonal piles are
made by placing loose pieces of bevelled wooden strips at each
comer of the form. The reinforcement is delivered on the work
in factory-made-units, so that it can be placed in the forms at
once. When a reinforcing unit is suspended and centered in
position, the concrete of a wet consistency is deposited and care-
fully puddled. As soon as the concrete of the pile has solidified,
the forms are stripped and used for making other piles. The
number of forms required will vary with the quantity of piles to
be made and the prospective salvage in the lumber.

The curing of concrete in the normal manner delays the
driving of the piles for a period of not less than 3 weeks, although
a greater length of time is desirable, especially in cool and damp
weather. Therefore, unless a stock of cured piles is always on
hand, it frequently occurs that this type of pile cannot be used
at all and resort is had to the use of the cast-dn-place type. This
practice is open to question on account of the inability of plain
concrete to resist even moderate tension. In fact, it is almost
axiomatic that all concrete piles must be reinforced. Every
concrete pile is subjected to strains that induce very serious
tensile stresses in the pile. Such stresses may result from super-
imposed loads or a lateral strain from the soil. Further, the

Digitized by

Google

318

CuMMiNGS ON Reinforced Concrete Piles.

making and storing of concrete piles for use at any time necessi-
tates a large financial investment.

In order to avoid the above-mentioned objections, the writer
has adopted the method of steam curing of cast-and-driven piles.
This enables such piles to be made and driven within 3 or 4 days
and places the speed of driving on the same basis as that of the
cast-in-place type.

The means used for steam curing will vary with circum-
stances, location of the work, speed required for delivery of the
piles, the number needed, etc. During the past winter the
writer has used concrete piles made and driven within 10 days,
and, as a result of this experience, confidently recommends that
such piles can be made, cured and driven immediately.

On the work with which the writer was connected, the piles

* — ^*^^..>

PJl ^ ^p^^ ^^-^^^^0^^

III

h -"iJ'-i?'"- -

_J J a D

\s>'^ '2> J •:

EN^Nt«4

FIG. 2. — SHED FOR STEAM-CURING PILES.

were allowed to set in a normal manner for 5 or 6 days — and
were then gently hoisted^ from**^ the molding bed by*a derrick,
using an equalizing spreader and bridle, the chains of the bridle
being fastened so that the pile was balanced. They were then
placed in stacks of 25 or 30 and separated from one another by
wooden blocks, particular attention being given to securing a
solid bearing for each pile.

A light wooden shed, practically steam tight, was built
entirely around the stack of piles. Fig. 2. The steam was con-
ducted direct from a boiler through a 1-in. pipe to 3 branch
openings inside the shed. The steam pipe valve was opened and
the piles were exposed to live steam for 2 or 3 days, when they
were found to be ready for driving. On being first exposed to
steam, the moisture condensed on the surface of the piles and
remained until absorbed by the concrete when the temperature of

Digitized by

Google

CUMMINGS ON ReINFOBCED CONCRETE PiLES. 319

the steam was reached. This steam treatment should be distin-
guished from heat applied indirectly or baking, in accelerating
the set and hardening the concrete. The writer sees no reason
why boiling water should not be used for the same purpose if
the conditions are favorable for adopting this method. Pre-
caution should be exercised in making sure that the concrete
has solidified and that it has received its initial set, before expo-
sure to steam treatment.

The prospective field opened up by the steam treatment
for the rapid curing of concrete seems to the writer to solve the
diflBculty incidental to the present methods of procedure in the
construction of all classes of concrete structures.

Attention is directed to the publication of the tests of the
Structural Materials Laboratory of the United States Geological
Survey, wherein it is conclusively shown "that a compressive
strength considerably (in some cases over 100 per cent) in excess
of that obtained normally after ageing for six months, may be
obtained in two days by using steam pressure for ciu*ing mortar.'*

Practicing engineers will have little difficulty in modifying
the writer's methods and using improved schemes for quick cur-
ing of the piles with steam; and in this connection it may be of
interest to state that the writer has already under way the con-
struction of steel molds and appliances for the steam treatment
of concrete piles.

A list of references on Concrete Piles to 1908 is given in the
appendix.

Digitized by

Google

APPENDIX.
L18T OF REFERENCES ON CONCRETE PILES.

Concrete Piles for Sandy Ground. Engineering News, v. 49, p. 275.
(March 26, 1903.) (Description of the Raymond system of sinking piles
with water jet. Illustrated.) (1 column.)

A Concrete-Steel Pile Foundation. Engineering News, v. 49, p. 173.
(February 19, 1903.) (Description of piles used for the Court House,
Berlin, Germany. Driven by steam hammers.) (1 column.)

Concrete Pile Foundations, Carnegie Public Library, Aurora, III.
Engineering News, v. 48, p. 495. (December 11, 1902.) (Description
of the Raymond method of using steel core for making holes for concrete
piles. Illustrated.) (§ column.)

A New System op Concrete Pile Construction. Engineering News, v.
45, p. 450. (June 20, 1901.) (Description of the trial of the Raymond
steel core system, made at Chicago, May 16, 1901. Illustrated.) (1
column.)

Concrete Pile Foundations. Engineering News, v. 46, p. 75. (August 1,
1903.) (Description of the foundations of the nine-story apartment house
built for W. J. Bryson, Lake Avenue, Chicago. Holes bored by water-
jet system before Raymond steel core was dropped. Material largely
sand and quicksand.) (| column.)

The Hennebique System of Armored Concrete Construction. By
Leopold Mensch. Journal AssodcUion of Engineering Societies, v. 29,
p. 108. (September, 1902.) (Contains three pages on concrete-steel
piles.)

Abstract of same. Concrete-Steel Piles of the Hennebique System.
Engineering Record, v. 46, p. 618. (December 27, 1902.)

Construction in Concrete and Reinforced Concrete. By F. C. Marsh,
1902. Minutes of Proceedings of the Institution of Civil Engineers,
V. 149, p. 297. (Gives a short description of piles made after the Henne-
bique System.)

Neuere Bauwesen und Bauwerke in Beton und Eisen, nach dem Stands
bei der Pariser Weltausstellunq 1900. Fritz v. Emperger, Zeit-
sckrifl des Oeslerreichischen IngenieW' und Architekten-Vereines, 53.
Jahrgang, pp. 713 and 765. (October 25 and November 15, 1901.)
(General description of concrete piles and their use.) (Illustrated. 7
pages.)

Abstract of same. Concrete-Steel Piles and their Driving. Engineering
Record, v. 46, p. 560. (December 13, 1902.) (Gives extracts from Mr.
C. F. Marsh's paper read before the Institution of Civil Engineers and
from Mr. F. Von Emperger's description of a pile driver from the Zeit-
schrift des Oeslerreichischen Ingenieur- und Architekten Vereines, Novem-
ber 7, 1902.) (2 columns.)

(320)

Digitized by

Google

CuMMiNGS ON Reinforced Concrete Piles. 321'

Concrete Piles. Railroad GazeUe, v. 34, p. 645. (August 15, 1902.) (De-
scription of the Raymond System with illustrations from the J. I. Case
Plow Works, Racine, Wis.) (1 column.)

The Raymond Concrete Piles. Cement and Engineering News, v. 13, p.
22. (August, 1902.) (Illustrated description of the Raymond System.)
(1 page.)

Concrete Pile Foundations op the Hallenbeck Building, New York.
Engineering Record, v. 47, p. 377. (April 11, 1903.) (Piles sunk by
means of water jet. Material, gravel and sand. Illustrated.) (1 page.)

QUAIMAUERN UND FuTTERMAUERN AUB BeTON UND EiSEN (SySTEM HeN-

nebique). Zeitschrift des Oesterreichischen Ingenieur- und Architekten"

Vereines, 53. Jahrgang, p. 539. (August 9, 1900.) (Short descriptions

of quay walls in Southampton, Paris and Nantes.) (Illustrated. H pages.)
Ueber Beton-Eisen-Piloten. Zeilachrift des Oesterreichischen Ingenieur-

und ArchUekien^Vereines, 54. Jahrgang, p. 746. (November 7, 1902.)

(Illustrated. 2 pages.)
Bbtoneisen-Ppahlrost vom Neubau des Amtsgerichtes-Weddinq in

Berlin. Deutsche Bauzeitung, 36. Jahrgang, p. 582. (November, 1902.)

(1 page.)
Pfahlrobtkonbtruktionen in Beton-Eisen. Deutsche Bauzeitung, 36.

Jahrgang, p. 411. (August, 1901.) (Short description of use of concrete

piles according to Hennebique system for a building for Holland-American

line in Rotterdam.) (i column.)
Beton-Pfbiler der SANGAMON-FLUSS-BRtJcKE. Thonindustrie Zeitung,

26. Jahrgang, 1. Halbjahr, p. 83. (January 17, 1901.) (Description of

the use of concrete piles for the St. Louis, Peoria and Northern Railroad

Bridge over the Sangamon River.) (f column.)

StaHL-BeTON-PfEILER Fto DIE CLYBOURN-PLACE-BRtJCKE IN CHICAGO, IlL.

Thoninditsirie-ZeUung, 25. Jahrgang, 1. Halbjahr, p. 631. (April, 1901.)
(Short description of steel-concrete piles for a drawbridge over Chicago
River.) (i column.)

EiNRAMMEN VON Betonpfahlen. ThonindiLstrie Zeitung, 27. Jahrgang, 1.
Halbjahr, p. 296. (February 21, 1903.) (Short description of method
of driving concrete piles.) (Illustrated.) (i column.)

Ueber Betonpfahle. Thonindustrie Zeitung, 27. Jahrgang, 1. Halbjahr,
p. 1106. (June 13, 1903.) (Short description of construction and
method of driving concrete piles.) (Illustrated.) (§ column.)

Recent Developments in Pneumatic Foundations for Buildings. By
D. A. Usina, Associate American Society Civil Engineers. Proceedings
American Society of CivU Engineers, v. 34, p. 220. (March, 1908.)
(Contains two pages on the comparison of concrete piles and caissons for
foundations of buildings.)

Foundations; an Informal Discussion at the Annual Convention,
July 10, 1907. Proceedings American Society Civil Engineers, v. 33, p.
812, 816. (September, 1907.) (Contains some data on concrete piles.)

CoNCBBTE and Concrete-Steel. Transactions American Society CivU
Engineers, v. 54, Pt. E., pp. 436, 461, 469, 548, 615. (International

Digitized by

Google

322 CuMMiNGs ON Reinforced Concrete Piles.

Engineering Congress, St. Louis, 1904.) (Very short references to
concrete piles.)

Cyundrical Foundations for a Quay Wall in the Harbor of Dblfzyl.
By W. F. Druyvesteyn. Transactions American Society Civil Engi-
neers^ V. 54, Ft. E. (International Engineering Ck)ngres8y St. Louis,
1904.) (Contains 1 page on concrete piles.)

Use of Reinforced Concrete in Buildings. By Frank C. Schmitz.
Proceedings Brooklyn Engineers^ Club, 1906, p. 67. Brooklyn, N. Y.
1907 . (Contains a short table comparing the price of wooden and concrete
piles.)

Driving Concrete Piles Below the Battery Tunnel, New York.
Engineering Record, v. 65, p. 678. (June 8, 1907.) (The construction
consists of a series of transverse pile bents at irregular intervaLs of about
50 feet supporting the undersides of the tubes for two lengths of about
600 feet each. In each bent there are two reinforced concrete piles 20
inches in diameter and about 7 feet apart on centers.)

Grade Correction and File Foundations in the East River Tunnel
OP THE New York Rapid Transit Subway. Engineering News, v. 57,
p. 718. (June 27, 1907.) (An illustrated description of the construction
of the reinforced concrete piles.)

Concrete Pile Foundations for a Tower 7(X) Feet High. Engineering
Record, v. 55, p. 531. (April 27, 1900.) (The piles are of the Raymond
concrete type and are made in the standard manner by ^t driving with
a solid steel core a thin conical steel shell, which excludes water and sand
and is filled with concrete after the core is withdrawn.)

Steamship Terminal with Concrete Pile Piers at Brunswick, Ga.,
Atlantic and Birmingham Railway. Engineering News, v. 56, p. 654.
(December 20, 1906.) (Gives specifications for concrete piles, illustra-
tions and method of construction.)

Cement Piers. Scientific American Supplement, v. 63, p. 26241. (May 25,
1907.) (The cement cylinders for use in piers at San Francisco are made
of three wooden piles enclosed in reinforced concrete.)

Improved System of Concrete Piling. Journal of tke Franklin Institute,
v. 160, p. 455. (December, 1905.) (A report of a committee on the
merits of the concrete pile invented by Frank Shuman, illustrated.)

The Simplex System op Concrete Piling. By Constantine Shuman.
Proceedings Engineers* Club of Philadelphia, v. 22, p. 347. (October,
1905.) (lUustrated.)

Concrete Piles, Description of the Methods of Manufacture and
Usages of the Two Leading Types op Concrete Piles which are
Replacing the Wooden Products. By David Lay. Cement Age,
V. 2, p. 626. (February, 1906.)

Reconstruction of the Atlantic City ** Steel Pier" in Reinforced
Concrete. Engineering News, v. 56, p. 90. (July 26, 1906.) (The
reinforced concrete piles were molded on small pile platforms adjacent
to the location of the piles in the piers; after hardening the piles were
lifted from the platforms, set in position and sunk into the sand by
means of a water jet, having a pressure of 65 lbs. per sq. in.)

Digitized by

Google

CuMMiNGs ON Reinforced Concrete Piles. 323

Cost of Making and Placing Reinforced Concrete Piles at Atlantic

City, N. J. Engineering News, v. 56, p. 252. (September 6, 1906.)
The Manufacture and Use of Concrete Piles. By Henry Longcope.

Proceedings National Association of Cement Users, v. 2, p. 277. (1906.)
Abstracts of same. Scientific American Supplement, v. 61, p. 25375. (May

12, 1906.) Municipal Engineering, v. 30, p. 106. (February, 1906.)
Reinforced Concrete Pile Foundation for the Lattemann Building,

Brooklyn, N. Y. Engineering News, v. 54, p. 594. (December 7,

1905.) (The piles used were of the corrugated form invented by Frank

B. Gilbreth.)
The Making and Driving of Corrugated Concrete Piles. By Frank

B. Gilbreth. Association of American Portland Cement Manufacturers,

Bulletin No. 7. (1906.) (Illustrated.)
L'Emploi des Pieux en B£ton pour les Fondations. Le Genie Civil, v.

49, p. 104. (June 16, 1906.) (An illustrated description of different.

types of concrete piles.)
Translation of same by George L. Fowler. The Use op Concrete Piles.

RaUroad Ga^tU, v. 41, p. 238. (September 21, 1906.)
Appontement M£tallique de Lome (Afrique occidentale). Le Genie

Civil, v. 47, p. 178. (July 15, 1905.) (Contains a description of the

concrete piles.)
Corrugated Concrete Foundation Piles. Engineering Record, v. 52, p.

548. (November 11, 1905.) (Describes the method of constructing

the foundations for the Lattemann Building. Brooklyn.)
Building and Machinery Foundations in Quicksand. Engineering Record,

V. 53, p. 248. (March 3, 1906.) (For the Knickerbocker Building,

New York City, the foundations of the column and wall piers consist

of clusters of tubular steel piles 12 in. in diameter and { in. thick sunk to

bed rock and filled with concrete.)
Concrete Piling. Scientific American, v. 90, p. 248. (March 26, 1904.)
Die GrI^ndung des Amtsgerichtsgebaudes auf dem Wedding in Berlin

MIT BetoneisenpfXhlen. By Hertel. Beton und Eisen, v. 2, p. 246.

(October, 1903.)
Reinforced Concrete Piling. By A. R. Galbraith. Proceedings, Incor-
porated AssocicUion of Municipal and County Engineers, v. 31, p. 356.

(1904-5.) Spon & Chamberlain, 123 Liberty Street, New York.
Abstract of same. European Reinforced Concrete Piles. Engineering

Record, v. 52, p. 99. (July 22, 1905.)
Concrete Piles at the United States Naval Academy. By Walter R.

Harper. Engineering Record, v. 51, p. 277. (March 4, 1905.) (Gives

the comparative cost of wood and concrete piles; test of concrete pile

and methods of construction.)
The Strength of Pile and Concrete Foundations. Engineering Record,

V. 50, p. 358. (September 24, 1904.) (Results of experiments made to

determine the adhesion of timber piles to concrete, when imbedded in

that material.)

Digitized by

Google

324 CuMMiNGS ON Reinforced Concrete Piles.

Concrete Pile J'oundations at Washington Barracks, D. C. By Captain
John Stephen Sewell. Engineering Record, v. 50, p. 360. (September
24, 1904.) (2 pages.)

Reinforced Concrete Piles with Enlarged Footings for Underpinning
A Building. By J. Albert Holmes. Engineering Record, v. 51, p. 567.
(June 16, 1904.) (Describes the construction of piles for the foundation
of a building in Boston, Mass.)

Des Applications du Ciment Arm£. By G. Liebeaux, Reime Generate
des Clieynins de Fer, v. 24, Pt. 2, p. 525. (December, 1901.) (Contains
illustrations of concrete piles for foundations.)

Notes on European Reinforced Concrete Structure. Engineering
Record, v. 51, p. 38. (January 14, 1905.) (Contains one column on
ferro-concrete piles.)

Reinforced Concrete. (Pt. 162, 375.) By Albert W. Buel and Charles
S. Hill. Ed. 2, N. Y., 1906. Engineering News Publishing Company,
220 Broadway. $5 net. (Contains data on concrete piles.)

Abstract of same. The Construction and Use of Concrete Steel Piles
IN Foundation Work. Engineering News, v. 51, p. 233. (March 10,
1904.)

Concrete-Steel Piles. Cement, v. 4, p. 16. (March, 1903.) (A descrip-
tion of heavy pile drivers designed for driving concrete-steel piles.)

A Treatise on Concrete Plain and Reinforced. (P. 477.) By Fred-
erick W. Taylor and Sanford E. Thompson, Associate Members, American
Society Civil Engineers, N. Y., 1905. John Wiley & Sons, 43 East
Nineteenth Street. $5.

Concrete and Reinforced Concrete Construction. (P. 428.) By
Homer A. Reid, Associate Member, American Society Civil Engineers,
N. Y. 1907. Myron C. Clark Publishing Company, 13 Park Row.
$5 net. (Gives data on concrete piles.)

Reinforced Concrete. (Pp. 41, 181, 200, 449.) By Charles F. Marsh,
Member American Society Civil Engineers, and William Dunn. Ed. 3,
London, 1906. Archibald Constable & Co., Ltd., 16 James Street,
Haymarket. $7 net. (Very short references.)

Cement and Concrete. (P. 485.) By Louis Carlton Sabin, Member Amer-
ican Society Civil Engineers. Ed. 2, New York, 1907. McGraw
Publishing Company, 239 West Thirty-ninth Street. $5. (Contains
2 pages on concrete piles.)

Concrete Steel. (P. 169.) By W. Noble Twelvetrees, New York, 1905.
Whittaker & Co., 64 Fifth Avenue. $1.90. (Contains 10 pages on con-
crete-steel piles.)

Reinforced Concrete Piles; Their Making and Driving. Cement,
V. 4, p. 331. (November, 1903.)

A New System op Concrete Piles. By W. P. Anderson, Engineering Record,
V. 50, p. 494. (October 22, 1904.) (Piling for the Dittman factory
building, Cincinnati, Ohio.)

Concrete Piles for Building Foundations. Engineering Record, v. 49,
p. 596. (May 7, 1904.) (2 columns.)

Digitized by

Google

CiJMMiNGs ON Reinforced Concrete Piles. 325

A Practical Treatise on Foundations. (Pp. 467, 477.) By W. M.
Patten, Ed. 2, New York, 1906. John Wiley & Sons. $5. (Contains
information on concrete piles.)

EiNE NEUERB KaIMAUER MIT ElSENBETON-PFAHLGRt^NDUNG. ZeUschrifi

fUr Baurvesen, v. 57, p. 550. (Pt. 10-12, 1907.) (On the reinforced
concrete piles for the foundation of a new quay wall at Dusseldorf.)
Methods and Cost of Driving Raymond Concrete Piles for a Building
Foundation. Engineering-CoTitracting. (February 13, 1907.) (Gives
figures of cost computed from records obtained in constructing the pile
foundations for a building in Salem, Mass.)

Kt^NSTLICHE BeFESTIGUNG DES BaUBODENS MITTEL8T SCHWEBENDER PILO-
TAGE. By Ottokar Stern. Beton und Eisen. (January, 1907.) (Calcu-
lations and dimensions for concrete piling.)

Concrete Piles. By Charles R. Gow. Journal Associated Engineering
Society. (October, 1907.) (Illustrated.)

The Simplex System of Concrete Piling. By Thomas MacKeller. Jour-
nal Associated Engineering Society. (October, 1907.)

PiLOTis "Simplex." Le Ciment. (December, 1907.) (Illustrated.)

Concrete Piles — Forms, Advantages and Cost as Compared with
Wooden Piles. By C. W. Gaylord. Proceedings, v. 5, 1909. National
Association of Cement Users. (21 pages. Illustrated.)

Digitized by

Google

THE HANDLING OF CONCRETE IN THE CONSTRUC-
TION OF THE PANAMA CANAL.

By S. B. Williamson.*

The estimated amount of concrete that will be used in the
construction of the Panama Canal aggregates 5,416,645 cu. yd. and
is distributed as indicated in Table I.

Table I. — Status op Concrete Work on the Panama Canal on
January 1, 1912.

Location.

Terminal Docks at Cristobal

Gatun Spillway Dam

Gatun Locks

Culebra Cut: Revetment

Pedro Miguel Ix>ck8

Miraflores Locks

Miraflores Spillway Dam

Terminal Docks at Balboa

Municipal Roservoirs. Dams, Power HoufM>,
Bridges, etc

Totol I

Amounts in Cubic Yards.

Placed.

To be Placed.

Total.

500

167.500

1.761.345

786."696
671.860

'6."535

22.000

51.800
57.580
238.895
400.000
111.993
840,876
75,000
330.065

62.300

225.080

2.000.240

400.000

892.689

1.412.736

75.000

336.600

22.000

3.310.436

2.106.209

6.416.645

While, no doubt, it is generally known that concrete plays a
leading part in the building of the Canal, a summary of the
quantities emphasizes its importance, which becomes still more
significant when one realizes that the estimated cost of the con-
crete structures represents 23 per cent of the entire estimate for
construction and engineering and that, had it been necessary to
adopt stone masonry, the cost of these structures would have
narrowly escaped a prohibitive figure, as there is no building stone
within a reasonable distance from the Canal Zone. It does not
seem out of the way, therefore, to advance the claim that the
lock type of canal, now under construction, and acknowledged
to be the most preferable by all engineers who have given the
subject careful consideration, became feasible largely through the
use of concrete.

* Engineer, Pacific Division. Panama.

(326)

Digitized by

Google

Williamson on Handling Concrete at Panama. 327

The locks comprise a large proportion of the total quantity
of masonry and in order to place the amount of concrete involved,
economically and at a rate to accomplish the work within the
allotted time, it was necessary to devise especial and unusual
handling appliances; and it is the purpose of the writer to describe
the plants and methods adopted.

GATUN LOCKS.

General Description. — ^At Gatun the difiference of 85 ft. between
sea-level and the lake surface will be overcome by a flight of three
locks, that is, the locks are directly connected, without inter-
vening basins, and form a continuous structure. They are built
in duphcate, which requires the construction of two side walls
and a center wall, each 79 ft. high. The side walls are 4,038 ft.
long; the center wall is extended in both directions to provide
guide walls for vessels entering the locks and is 6,330 ft. in length.
An outline plan and typical cross-sections of the Gatun flight of
locks is shown on Pig. 1, Plate I, and 2,000,240 cu. yd. of
concrete will be used in their construction.

The lock sites are about 3,300 ft. east of a channel that was
dredged by the French Company from Colon to Gatun, and as
the concrete aggregates were to be obtained from points on the
coast and transported by water to Colon, it was considered advis-
able to continue this method of transportation to Gatun by utiliz-
ing the French Canal; the latter, therefore, at once became a
controlling element in designing the handling plant. The banks
of the French Canal are composed of an alluvial material, with
a decided tendency to slide, and were considered unsafe for the
heavy structures and storage piles required at the unloading point.
A boat slip was therefore dredged in firmer ground between the
canal and lock sites and a channel excavated to a connection with
the French Canal — ^incidentally the movement of the unloading
point shortened the distance between the storage piles and mixing
plant.

Handling Plant. — Referring to Figs. 2 and 3, Plate II, it is seen
that the entire plant is composed of the following units, each
having a separate and distinct function:

1. Facilities for unloading and storing cement, sand and stone.

Digitized by

Google

346 Williamson on Handling Concrete at Panama.

Berm Cranes, — Two berm cranes only were used at Pedro
Miguel, but the booms were replaced by the cantilever arms of the
other two so that, as erected, they were balanced cantilever cranes
as shown in Fig. 13, Plate VII. Their runway tracks were so located
between the trestles that each cantilever extended well over the
respective storage piles; both cantilevers were equipped with trolleys
operating a 2i-cu. yd. excavating bucket. The bins from the other
cranes were also placed on these two, and with these and two
cantilevers the facilities for handling and storing sand and stone
were doubled. Aside from the above changes the previous

FIG. 12. — MIXING CRANES AND STORAGE TRESTLES, PEDRO MIGUEL LOCKS.

description of these cranes applies, except also that the swinging
platforms were omitted and the mixers emptied into buckets on
cars.

Narrow-Gauge Road. — The track system of the narrow-gauge
road used for transporting concrete from the berm to chamber
cranes is shown in Fig. 11. The difference of 30 ft. in elevation
between the forebay and lock floors was overcome by means of an
inclined trestle having a 2.5 per cent grade. The tracks were laid
with 70-lb. rails which enabled the locomotives to attain a greater
rate of speed than would be safe on the lighter rails usually
employed. The equipment included twelve llj-ton Porter

Digitized by

Google

Plate VII.

Proc. Nat. Assn. Cement Users.

Vol. VIII, 1912.

Williamson on the Handling of Concrete
IN THE Construction of the Panama Canal.

Digitized by

Google

1 I Miiffi

Digitized by

Google

Williamson on Handling Concrete at Panama. 347

locomotives and 24 steel-framed flat cars, all equipped with air,
each car being large enough to hold two 2-cu. yd. buckets.

The trains were composed of 2 cars and each car carried a
bucket, so placed that when alongside a berm crane each bucket
was filled from the corresponding mixer without moving the train.
Usually the trains alternated in going into the respective lock
chambers and stopped under the first chamber crane; the crane
placed an empty bucket and picked up a loaded one from the
same car; the train then moved to the next chamber crane where
the operation of exchanging an empty bucket for a loaded one

FIG. 14. — GENERAL VIEW OF LOCKS AND CHAMBER CRANES, PEDRO MIGUEL

LOCKS.

was repeated, after which the train of empty buckets returned to
the mixers.

Chamber Cranes. — Two chamber cranes, Fig. 14, were erected
in each lock at Pedro Miguel — they placed the concrete in both
side and center walls. Except that the long cantilever arms
extended over the side walls and the short ones over the
center wall, the description of them given for the Miraflores plant
applies.

Concrete Forms, — The same type of forms was used here and
at Miraflores, in fact, all of the steel forming and some of the wood
forms were used fiirst at Pedro Miguel and later at Miraflores.

Digitized by

Google

348 Williamson on Handling Concrete at Panama.

Auxiliary Mixing Plant, — Delays in the delivery and erection
of the permanent plant, combined with the desire to increase
the rate of placing concrete led to the erection of auxiliary mixing
plants for the Pacific locks. At Pedro Miguel a 2-cu. yd. mixer
was first temporarily set up at the lower end of the west wall and
later ipoved to a similar position as regards the east wall, where it
continued work until December 16, 1911. Two 2-cu. yd. cube
mixers were installed under the south end of the storage trestle
in the forebay and have continued in operation since. In all of
the above cases the mixers were on hand, having been ordered for
the permanent plant as applied to Miraflores where 8 mixers are
required as against 4 in the application of the plant .to Pedro
Miguel. In each location the mixers were charged from bins
to which material was delivered from tracks overhead, by standard
railroad dump cars, and the product was transported to the walls
by narrow-gauge equipment. Half-yard portable mixers are also
used for floor construction and certain portions of the walls where
they may be set up so as to pour concrete directly into the forms.

The stationary mixers located in the east wall at Miraflores
have been previously described. As they are for the purpose of
feeding the chamber cranes, and it was necessary to purchase
additional mixers, they really constitute an addition to the per-
manent plant, made for the purpose of increasing its efficiency.
Aside from these there is no auxiliary plant used at Miraflores,
except the half-yard portable mixers.

PERFORMANCE OF PLANT.

For the Pacific locks, crushed stone is delivered into cars
directly from the quarry bins and dumped from the cars on the
storage trestles. Sand is dredged, loaded into barges and trans-
ferred from the latter to bins at Balboa with electric cranes, it
then flows by gravity from the bins into cars for transporting to
the storage trestles. There is nothing corresponding to the unload-
ing plant at Gatun, therefore, unless it is the crane for handling
sand at Balboa. Their performance is given in Table VII.

Pedro Miguel Locks.

Auxiliary Plant, — The placing of concrete in the lower guide
wall at Pedro Miguel began on September 1, 1909, with a mixing

Digitized by

Google

Williamson on Handling Concrete at Panama. 349

S
>

Q
o

CO g

a g

o -I a

a
eg I

8 ^

<£

1^^

.s »

ooeo«oaaoo»oo^o»^o

t^f^cooo6h«t^c0(Ot^tov}

iQe>«iQe>«i-HQot^^QO-«co

sggsss

g^^^r^ss^g^^^^^ss

8$SS88S8SS88

Q«OQ«O)Ci-iC0Q0C

3:22§gS8S88888

8S8 : : : :8 :

S2«S8SS888888

O u

t«

32^888

8388SS88 : :88

o>-«oa>raoc<4eM • -e^oo

g^^8S8S88888

S55

5 <^ .-4 ic .^ oi X S

e<)C9NC4e4c^oic4MMo«e4 i n

• a

o S
HO*

ao90 0o>o^^

S3$$3$SS

8888888

SSS8888

8888888

-«Q0

8 8

88S8888

ssi^g'sii

C4C^^C4OlO«04 1,^ I

Hum I !

Digitized by

Google

350 Williamson on Handling Concrete at Panama.

P
C

o I.

O »-

►< &

< ^

<
o

I
I

o'^

1:^

^°

Sao

•Sfi

aiaa

^00

.T3

> a

:8S8S8S888S888S88S88

"Sri

:Si:iS3!;S!^SSSSSSSSi;S&SS

:2SgJ2SS8SSS?SJ5S2GS2S

:$S3i:S3;3$:;S33S:8S3SSS

;assss|s|sgg§g|§Sgg|

g> t-" o « -* i-i lO d d e<i « ci •* CO i-J eo CO

S88SS8S8§8§S8§888

5 ^ « »o c« t^ »H « ^ M ts. oj 00 55 ti cft <©

• ir)t<idoddtxidoddod»^odh<iad*oi»(0<ot>ii

°s3sssssssss;ss;sss;s3!;

— I

. ■QQQ«OQtxC<ICOaQC<)'HP|<DeoOOOe>«Q

• ■ooo^o^i-ir*oO'HO«*o>'^«>o«o

I I I § § E V .-^i

00 ^ S iS '4* N> a» ? ra CI S S S o5 o r» <o

8 8 S S ^ 3 S S m 8 ? tx S lO Ss o 00

SSS|2:S|S§i§;2l88gi:iS

S8S{2f:S82222g5^825?

«a>-«ec>o^ic>oc4»or^iO>o>ctoioio

S^*ss§s5s:55;ssssss5t:

:f :i! ::R :« :?;

C4NNCIC4C4e<)C4 CI CI 09 04

QQQO4QO4t»-«Q00QQt^tOOia0a»

^ ,^ ,^ ^e>io) ^04 f-ioi ^o< ^ ^ i-ioi ^

^l?;b

Digitized by

Google

Williamson on Handling Concrete at Panama. 351

mil

III

gnssssj;

S«SS!oSSS

•coO'-iwo*^©

c5«?3c5?iSn

■ ■S S «^ S c ^

ilis ii I

Digitized by

Google

352 Williamson on Handling Concrete at Panama.

plant consisting of two 2-cu. yd. and three i-cu. yd. mixers. The
performance of this equipment to Jmie 30, 1910, working on an
8-hr. day basis, is detailed in Part 1 of Table VIII.

During the fiscal year ending June 30, 1911, the auxiliary
plant consisted mainly of three 2-cu. yd. mixers, one located at
the south end of the east wall and two in the forebay, the half-
yard having been transferred to Miraflores. It placed 121,530
cu. yd. and the detailed performance, on an 8-hr. day basis, is
shown in Part 2 of Table VIII; and the detailed performance,
on an 8-hr. day basis, of this plant for the first seven months of
the fiscal year ending June 30, 1912, is shown in Part 3 of the same
table.

Permanent Plant, — One berm and two chamber cranes, or
one-half of the plant, began operating on April 4, 1910 (a chamber
crane placed some concrete from auxiliary mixers in March);
the other half began on July 15, 1910. The portion of the per-
manent plant in operation laid 73,083 cu. yd. prior to June 30,

1910, on an 8-hr. day basis, as detailed in Table IX.

The plant worked as a whole from July 15, 1910, to January
31, 1911. On the latter date the dismantling of one berm crane
began preparatory to erecting it at Miraflores. The dismantling
of two of the chamber cranes for the same purpose began on
April 20 and May 9, 1911, respectively, and that of the remaining
berm crane on May 19, 1911. Two chamber cranes remained at
Pedro Miguel until December 12, 1911, and January 31, 1912,
respectively, being used in the meantime for placing concrete
from auxiliary mixers, setting iron work and backfilling the middle
wall. The plant placed 379,190 cubic yards during the fiscal year

1911, and the detailed performances of the berm and chamber
cranes respectively are given in Tables X and XI.

Miraflores Locks,

AiLCiliary Plant. — Placing concrete in the floors and lateral
culverts of the upper locks at Miraflores was begun on June 1,
1910, with a plant consisting of two ^-cu. yd. mixers, and 1630 cu.
yd. were placed before June 30, 1910.

In the fiscal year ending June 30, 1911, the auxiliary plant

Digitized by

Google

Williamson on Handling Concrete at Panama. 353

525
O

S
>

Z K

5
2

3
o

l«

•s .

•-1 00 uSa

:s§^g;

illtii

ss3;s:ss;^SoSSS|s^

OQ 04* to 00 t*' CO ^' CO CO 9 1^

a lO lO 55 »o ■* «o iS -* CO »-i I

g :

iQ CD e^ Q a» CI CO CO :«> 00 Q

oSS'^'-iioSSoSo

^Neoeocieocoodcowco
t^O>0<0-H00<-*C4C0S
ooa»oo»^io^to^ra

88
§8

(Ooaoeoo^cD^coo
SSS«^oQ>oa»Scot*

^'codi-Io6ci«o«o2i> •

S^SBSiiSSSSSSS

Nor^iHio^Neotoc

iOr^iooot*cocooo«

oeo

S M ra^9»^ CO S 0(000

oi :* lO CO »0 00 N to CO £1 -^

Vicococit>^r<^'-4«p«dto
N.3Sco>*'*eo»cSir^

?25

Concrete
Mixed,
cu. yd.

34.890
50.808
49.852
55.152
49.568
31.156
35.640
23.770
25.704
17.188
5.462

8 :

Number

Mixere.

(2ou.yd.)

888S8t3Se8888 § :

^'^^eo'^eocic4 0«C4e<« co •

I 1

888S8?^^8888|S

ltlilii-»

iill^ 11

Digitized by

Google

354 Williamson on Handling Concrete at Panama.

55
O

^ g

s s

« o

5 s
(« s

8 I

Raht

UN
11 |S

|.2l

oj? o S 5,0

o
H

10 ^ CP <«' <0 (0 to A 00 01 94 -<
00Cnt^t«t«iO«OtO<AtCiQ<4'

(D t» r» t^ « o SS to 10 10 iS

xot*coeo

r«'^9t0C4aa9ixoS^i-

QDC0C4t<«-<a»tO^tOt»t>««

:SSS255S23gS

•O CO

^^ lO c« N N .-• -^ ^

Dt<«CPt»(O^tO'^t0C0<-<'-4

9 5

.Sg

Slo^-^aoS-^owN-eo o 5

Dt^«t^»^t0^iB«i-i,-4

^11

£J

ooOQOcoeor'.MCooNgo^'

O)kO00«OiO^QOli-iQWC0

S8SS8SSS8^^§§|^

CO <« 09 09 ^ CO CO ^ ^ CO -4 f-i I w

«o d to ci CO »^ CO

ooieooe^to^

^^'^•tO^COM

^S:;:S^S8

tOtOQQ

ocoSt»tora^

S^S8S^S

e«^®Q^Nto
t^eoototo^pi

»-i-^C5N»-»i^

s:§SSS^S^

SS888^$iS

^itlfli I

-a

►

Digitized by

Google

WlLUAMSON ON HANDLING CONCBSTE AT PANAMA. 355

consisted of two 2-cu. yd. mixers located under the north, end of
the east storage trestle, and four i-cu. yd. mixers. The 2-cu. yd.
mixer plant was moved to the east wall for supplying the chambers
cranes in May, 1911: previous to this it supplied a berm crane
that was sufficiently complete to place concrete with the boom,
though its mixers and cantilever arm were still in use at Pedro
Miguel. The auxiliary plant handled 205,255 cu. yd. during the
year, as detailed in Table XII; and a similar statement giving the
detailed performance of the 2-cu. yd. auxiliary mixers at Mira-
flores, on an 8-hr. day basis, for the first seven months of the fiscal
year ending June 30, 1912, appears in Table XIII.

Permanent Plant. — One berm crane without cantilever arm
and mixers began, in its uncompleted condition, to place concrete
supplied by the auxiliary mixers on September 2, 1910, and
continued placing until February 15, 1911, when it was taken out
of commission for completion. It began operating again, as a
complete machine, on March 22, 1911, and a second berm crane
began work on April 7, 1911. With these 67,774 cu. yd. were laid
previous to June 30, 1911.

Other units of the plant were placed in commission as follows:
2 chamber cranes on July 13 and August 3, 1911, respectively,
and 2 additional berm cranes on July 25 and October 28, 1911,
respectively. There are 2 more chamber cranes under erection as
follows: Table XIV shows the performance of berm cranes,
Miraflores Locks, to January 31, 1912, and Table XV the perform-
ance of chamber cranes, Miraflores Locks, for first peven months
of fiscal year ending June 30, 1912.

The performances of the Pacific Division plants for the fiscal
year 1911 are given by the Cost Accountant as follows: In Pedro
Miguel locks 497,802 cu. yd. of concrete were placed, average
division cost $4.70 per cu. yd. and 385 cu. yd. of reinforced con-
crete at $17.74 or a total of 498,187 cu. yd. at $4.71 per cu. yd.
In Miraflores locks 272,933 cu. yd. were laid at an average
division cost of $4.68 per cu. yd. The lowest average cost for
any one month was for Pedro Miguel in November, 1910,
when 64,248 cu. yd. were placed at $4.20 a cu. yd. and for Mira-
flores in May, 1911, when 36,154 cu. yd. were placed at $4.05 per
cu. yd.

Digitized by

Google

366 Williamson on Handling Concrete at Panama.

>
Q
u

2 -

S s

;^ 2

00 J

! 0^

b 5

QqO

«e^®--'fM»odM«M.-<ddh*gd^ddx-HMgo-^fdt«^diO--dec^5

«S5««c©o>wicoswr-'-<o6pi««.So^ioi««.^w?iNr*'*^e5NO»PO
bJt^ts!h«iio«odeod^«t«rao>ciodd^ddcoi^ice>4t^t^c>9io<d<o

a 3

:j::^:?::^ :f:j! :^ :?::« :f!:R :s:s :s:« :j!:f :s:f :j;xj!

C4 C4 04 C4 es 04 d 04 «

Ill

^ I i I

5* -2 JJ 4s
»^ < 00 O

••2

Digitized by

Google

Williamson on Handling Concrete at Panama. 357

Digitized by

Google

368 Williamson on Handling Concrete at Panama.

>
Q

M
U

g a

" I

It. ^
S &

ri^lij^lll^

1-^

^6sB

SS8S

<Oaot^O

SSSJIS

n
.a

Is*

8 8

: :S :

S S

ii-

:2 :a '°g ^

8 to COM

dSSS

^>ouS

lulil^i

iilli

25 8

c««eo<o
Mcoooeo

€4 C4

8S36S

cieoeoeo eo

OooSS S

63.55
59.87
51.20
56.56
50.17
66.69
67.06

945.00 49.40 26.80
1,278.00 49.70 26.40
1,188.00 44.70 21.30
1,368.00 30.70 19.10
1.672.0029.70 12.80
1,575.34133.50 19.80
1,742.25 38.10 21.30

512.68
678.85
566.70
706.28
653.62
929.85
971.54

604.00
729.00
648.00
711.00
864.00
949.00
1,063.00

183.68
292.51
316.25
316.99
429.50
316.15
350.26

30.00
56.60
46.60
45.84
40.46
63.03
83.46

33.43
67.36
62.87
75.17
78.55
62.19
73.26

61.17
97.65
158.84
148.34
262.75
136.32
147.36

^
M

:::::::

''-

5.60
4.00
3.00
1.00
2.60
6.92
4.18

320.321 53.58
436.49' 67.90
331.75 44.94
395.01 45.64
434.50 45.26
632.85 47.69
712.74 42.00

273.45
387.21
309.07
367.09
373.60
560.15
692.76

26,326
33,726
26,349
26,200
19.427
31,180
37,054

^' >o to id r«' «D «D

2^8Sg3888

N eo c«' CO ^ ^ ^

July

August....
September.
October...
November.
December.
January...

Digitized by

Google

WiLUAMSON ON HaNSUNQ CONCRETE AT PaNABIA. 359

e
<2

S S5

§ 8

h g

>
«

•^e

p< o

aooooooot«>oio 1 1^

SSt^h-S^o

88SS8S^8

8 : :8 :5;S

S 3

S !<• 00 a> 00 a> to

S.3 3 I

^Op'*« »««-"<

$8S8mSS

«D<^CD^tOQ

M O «C O C«l 00 -^

•s^sss

sssssssm

III

Digitized by

Google

358 Williamson on Handling Concrete at Panama.

>
Q

«
U

^ .

M O

'^ I
o &

■Si

M
I

ssss

:;SSS

2SS2

2^SS

sssss

;;^55

^oas

'888S

3l
^2

cm' t>^ I— d

8 8

i §

^1^1

S S

2 'Sl]9j^

11^
ass

li"s

•5 us

sqi^s

C9t00<0

*^'2SS

8336^

esicoeoeo eo

8SS88 S

•H«-IMC^ *^

Jtii il^

•Q 00 S to *-< S O

00 ^ ro ^ 00 00 w

945.0049.40
1,278.00 49.70
1,188.00,44.70
1,368.00'30.70
1,572.00 29.70
l.S?^ 34 33.50
1.742 25 38.10
1

512.68
678.85
566.70
706.28
653.62
929 H5
B7i M

CD 1

183.68, 504.00
292.51' 729.00

316.25 648.00
315. 99| 711.00
429.50 864.00
316.15 949 on

350.26 1,063 jwi

rt iS * !^ I-, rs PS

I-' h- SC< 1^' -^1 ts t^

i£, a k^ -* «5 rt ^

S888SSS2

MS '^ CO "-• ci w ^
ioa33e4«8

8' CD ^ to ^' esl ci
CO CO a» CO CO •-)
CO ^ CO CO ■^ «o tN.

ooevi«a8r»S
«o o» ci i>i -i,^ ci ot

^ ^ CS| d «0 l>. 1-1

•«•• es o S i35 ^ t-

0COCOU»M3

gsi

^' u3 id to 1^ «c> «d lo

^SSS888|P^

In-^i^

Digitized by

Google

WiLUAMSON ON HaNSUNQ CONCRETE AT PaNABIA. 359

^ o

111'-

;$s

SSSS3$§

SsoSocoS

8 : :8 :^S S

*4 I le* •d«o

■4 a> CO <-• oo M CO

*-4«D 'CDiOi

:2g5S

•5 w. 9-2

SSSS8SS

•i 1

Sa«

^'StSS

lijroi

<"z

sssssss I s

■Mm

Digitized by

Google

360 Williamson on Handling Concrete at Panama.

The detailed division cost for the year is given for Pedro
Miguel as follows:

Concrete (495,037 cu. yd.):

Cement $1 .5365

Stone 8242

Sand 3729

Mixing 1771

Total cost of concrete $2.9107

Large rock (2,765 cu. yd.) $1 . 1483

Masonry (497,802 cu. yd.):

Concrete $2.8945

Large rock 0064

Forms 4387

Placing 3118

Reinforcements 0367

Pumps 0343

Power 0454

Maintenance of equipment 1723

Plant arbitrary 6847

Division expense .0792

Total division cost $4.7040

The division costs per month since June 30, 1911, have been
as follows:

Cost
Pedro I

per Cubic Yard of Concrete.

Month. 1911.

^iigucl.

Miraflores.

Plain.

Reinforced.

Plain.

Reinforced.

Julv

$5.82
5.63
6.08
6.26
5.80
6.27

$6.26
8.74

11.91
8.85
9.94
8.98

$4.93
4.45
4.41
4.50
4.89
5.06

August r r T T . T

$11.17

September

October

16.32
21 39

November

December

23.76
14.45

It is obvious from the tables that a large percentage of delays
in placing concrete at both Gatun and the Pacific locks is charge-
able to forms. At times the forms are filled so rapidly that it is
difficult to keep them ahead of the placing, but the greatest loss
of time is occasioned by the amount and complication of forms for
the electrical tunnels, conduits and machinery rooms near the
tops of the walls. The records given for the Miraflores plant are
not representative, as the entire plant is not in operation.

Digitized by

Google

USE OF CONCRETE IN THE FOURTH AVENUE
SUBWAY.

By Frederick C. Noble.*

The Fourth Avenue Subway, Brookljna, is an example of the
use of remforced concrete on a large scale. The portion now
under construction extends about four miles from the Brooklyn
end of the new Manhattan Bridge, over which it is intended to
connect with a subway system in Manhattan. The route, Fig.
1, lies through Flatbush Avenue, Fulton Street, Ashland Place
and Fourth Avenue, to Forty-third Street; in a direction generally
south. From here it is proposed to extend it in future by two
branches to the southern limits of the borough. The subway
is being built for the City of New York, under the supervision
of a state commission, with Mr. Alfred Craven as the chief
engineer. The work was divided into six contract sections, which
were let in November, 1909, at an aggregate price of about $15,-
000,000. Construction is now nearly finished.

The structure normally has space for four tracks; two for
local and two for express service. These are increased to seven
and eight in places where connection spurs, for future extensions,
are provided in two levels to avoid turnouts at grade. There
are six local and two express stations, with platforms long enough
to accommodate ten-ear trains.

The excavation was principally in a moraineal deposit, con-
sisting of sand with more or less gravel and boulders. No rock
was found anywhere on the line. Some quicksand was met
near the middle of the route, where the subway traverses ground
filled in over the bed of an old salt marsh. Ground water was
encoimtered at or near tide level and was controlled by pumping.
Excavation was usually carried on under covered roadways, but
in the extension of Flatbush Avenue open excavation was per-
mitted.

In connection with the work it was necessary to underpin

* Division Engineer. Public Service CommiMion, Broolclyn. N. Y.

(361)

Digitized by

Google

362 Noble on Use of Concrete in Brooklyn Subway.

about forty buildings with masonry piers to sub-grade in advance
of the excavation, where it approached close to and below their
foundations. It was necessary also to temporarily support the
elevated railway on Fulton Street and to drift a crossing under
the present subway in operation in Flatbush Avenue. Since
Fourth Avenue lies along the foot of a considerable drainage
area, sloping from Prospect Park, it was necessary to depress or
intercept many cross-sewers carrying a heavy storm-flow, and
to provide under-grade crossings at intervals. About six miles

Fia. 1. — PLAN OF FOURTH AVENUE SUBWAY, BROOKLYN, N. Y.

of sewers, of all sizes up to 9i ft. diameter, were thus built or
rebuilt.

The typical four-track section is shown by Fig. 2. The roof,
sides, intermediate walls and floor, where below ground water,
are reinforced with bars ranging between li and f in. square.
The design provides for a uniform live load of 300 lb. per sq. ft.
at the street surface. Waterproofing is used at stations on the
roof and sides and generally below ground water on the sides and
floor. Elsewhere the concrete is relied on for sufficient protec-
tion against seepage.

The amount of concrete required is somewhat over 400,000

Digitized by

Google

Noble on Use op Concrete in Brooklyn Subway. 363

cu. yd. The usual proportions are 1 to 2J to 4J. Much of the
excavated sand is suitable for concrete and is used accordingly.
The remainder is washed or dredged sand from the north shore
of Long Island. The aggregate is mostly washed or dredged
gravel from the same source, usually under 1 in. and well graded
in size. At times when gravel is not readily obtainable, cobbles
and fragments of boulders are crushed and used. The concrete
is mixed rather wet to facilitate its flowing around the reinforce-
ment, as this is generally spaced quite close.

Reinforcing rods were required to be deformed. Trans-

OMiriM IOO^Iam

SMtioii Ml DryGrMBd.

TkiekMSS •! concrete and spocin^
of rods ore ^ivtn for nlnimwiB cover, ond
ciiaajo whort cofor aCMtfo 7 ft.

FIG. 2. — TYPICAL POUR-TRACK SECTION OP SUBWAY.

verse and longitudinal rods were wired at their intersections
(Fig. 8). For this purpose, special devices of bent wire were
used; these served at the same time to space the rods, or to hold
them at the proper minimum distance from the forms. Where
imusually long spans or concentrated loads occurred, the roof
was of girder and jack-arch construction. To hold the thin
covering of concrete imder the lower flanges, special clips or
hangers of twisted wire were attached at intervals by bending
them over the upper sides of the flanges. Sometimes a strip of
coarse wire mesh was wrapped loosely around the flanges to
serve the same purpose. In a few instances, where the concrete

Digitized by

Google

364 Noble on Use of Concrete in Brooklyn Subway.

under the flanges tapped hollow, it was chipped off around the
edges of the flange, and a strip of wire mesh was attached and
covered with mortar blown in place by means of a cement gun.

Concreting was permitted throughout the winter, except
on the very coldest days. Materials were heated by fires or
steam coils, and the freshly laid concrete was covered with salt-
marsh hay and tarpaulins. Salamanders were placed under the
roof forms in some places. On accoimt of the varying condi-
tions of mixing and degree of exposure, it was impracticable to
devise a set of rules to fit all cases; but the criterion was held to
be the temperature at which the mixture could be deposited in
the forms.

Roof forms were struck as soon as the temperature condi-
tions and setting qualities of the particular brand of cement
used would permit. As this was a factor limiting progress, espe-
cially in the case of large steel forms, it was desirable to strike
them as soon as practicable. In summer they were struck after
80 hours, and in exceptional cases even at 50 hours, but in winter
it was sometimes necessary to wait a week or 10 days before
striking.

The methods of mixing and handling concrete vary between
the diffei;pnt sections and are described briefly as follows:

Section Af. B. Ex, i, extending from Nassau Street through
Flatbush Avenue to Willoughby Street, is being built by Smith,
Scott & Co., contractors.

Materials for concrete are brought on the work over a tram-
way extending along each side of the cut from a dock \mder the
Manhattan Bridge. Sand and gravel are loaded at the dock in
side-dump cars, drawn by 20-ton dinky locomotives. The cars
are run up an incline and dumped into bins over the mixer, which
is located about centrally on the section. This is a li-yd. machine,
operated by a 30-h.p. electric motor. The mixture is discharged
at the track level into bottom-dump cars of 1-yd. capacity, in
trains of 3 or 4 cars, hauled to the point of deposit and dumped
in frame chutes through which it slides to the forms.

As the sides of the cut were almost self-sustaining and as
the contract terms allowed open excavation on this section, very
little' cross-timbering was necessary; a condition that greatly
facilitated the erecting and moving of forms. The forms were

Digitized by

Google

Noble on Use of Concrete in Brooklyn Subway. 365

of the collapsing pattern, with stiffened steel sides and wood top
to suit the varying widths of roof. The floor having been laid
for the entire width of structure, and the reinforcement placed
for the sides and intermediate walls, the forms were moved into
place, one 20-ft. section at a time, traveling on tracks laid on the
floor. In this way it was possible to lay concrete for the walls
and roof over four tracks in sections 20 ft. long at a continuous
operation. These forms were very similar to those used on one
of the lower sections, in which connection they will be more fully
described.

Where structural roof framing was substituted for rod con-

no. 3. — COLLAPSIBLE JACK ARCH FORMS.

struction, centering (Fig. 3) was used to turn the jack arches.
These were of angle-stiffened plate, with an adjustable center
strip of wood for varying the span, and were set in position on
falsework. The span was maintained by tumbuckle ties, which
also served to draw in the sides on striking the forms. After
striking, each section was lowered with a winch.

Section Q-C-l, extending from Willoughby Street along
Flatbush Avenue and Fulton Street to Ashland Place, is being
built by William Bradley, contractor.

Materials for concrete are delivered at a large yard along-
side the subway between Third and Sixth Streets, where the
contractor has a cement storehouse and a stable for his horses.

Digitized by

Google

366 Noble on Use of Concrete in Brooklyn Subway.

Teams are employed for moving all materials. Concrete was
mixed principally by a gravity mixer located in the cut. Storage
bins for sand and gravel were placed just below the street level.
These fed 4 loading hoppers, each of one-bag-batch capacity.
After moistening, the contents were dropped successively through
three mixing hoppers, receiving the full dose of water in the first
one. The middle hopper was slightly offset from the other two.
The mix was discharged at the bottom level into cars which
were nm onto a cage, hoisted above the street surface, and dumped
into an overhead bin. From the bin it was almost immediately
discharged through a gate into tight rear-dumping steel carts
of about If-yd. capacity, which were teamed to the point of
deposit. The mixture was dumped on hopper platforms and
conveyed through 8-in. telescopic chutes to the forms. The
opening at the bottom of the hopper was closed by a spatula until
the entire load was cleared.

Most of the structmre on this section is of steel bent and
jack-arch construction, so that no unusual form-work was re-
quired. The reversion to this older type of design was at the
request of the contractor, and partly because of the concentrated
loadings brought by the elevated structure and building foimda-
tions to be supported permanently on the roof.

Sections 11-E-l and 11-A-l, which together form one con-
tract section, extend from Fulton Street, through Ashland Place
and Fourth Avenue, to Sackett Street, and are also under con-
tract with William Bradley.

The methods of mixing and placing concrete are the same
as those described for this contractor's adjoining section. The
form of construction is also similar, except where the line passes
by a curve under the present subway in Flatbush Avenue. Here
the design is of massive concrete arches, one for each track,
without reinforcement (Fig. 4).

Section ll-A-^, from Sackett Street along Fourth Avenue
to Tenth Street, is being built by the E. E. Smith Contracting
Company.

Materials for concrete are stored in a yard beside the Gowanus
Canal at Third Street, and are distributed to the work by mule
teams. Concrete is mixed in the roadway, directly over the
point of deposit, in five small readily-shifted drum mixers of

Digitized by

Google

Noble on Use of Concrete in Brooklyn Subway. 367

i-yd. capacity, each operated by a 15 to 25-h.p. electric motor.
The mixture was discharged and conveyed to the forms through
10-in. sectional chutes of 12-gauge iron, with a funnel at the top.

Generally, the floor and walls of the westerly two tracks
were built first, then the floor and walls of the easterly two; after
which the roof was carried across all foiu- tracks in sections about
30 ft. long. Five such sections would be in progress at one time.

The necessity for maintaining the cross-bracing of the trench
precluded the use of large forms such as were used on certain of
the other sections. Forms were made of 2-in. lumber, dressed
all sides, and with dapped edges. They were assembled in well-

FIG. 4. — TYPICAL SECTION THROUGH SUBWAY AT TUNNEL UNDER PLATBUSH

AVENUE.

braced panels and could be used many times. Jack-arch forms
(Fig. 5) were made of 16-gauge iron, nailed on wood ribs and
were suspended in position by long bolts.

On this section the usual waterproofing on sides and bottom
below ground water level is omitted. In such situations the con-
crete is locally made richer, 1-2-4, and the longitudinal rein-
forcement is doubled (Fig. 2).

Section 11- AS, from Tenth Street along Fourth Avenue
to Twenty-seventh Street, is under contract with the Tidewater
Building Company and Thomas B. Bryson.

Materials for concrete are brought on scows to a dock at

Digitized by

Google

368 Noble on Use of Concrete in Brooklyn Subway.

the foot of Nineteenth Street, from which a 3-ft. tramway leads
to the cut and through it in both directions. The sand and
gravel are transferred from the scows to overhead bins by a crane
and clam-shell bucket. The bins discharge through gates into
4-yd. side-dump cars. These are hauled by a 20-ton locomotive
to the main storage bins, of about 3,000 cu. yd. capacity, situated
in the cut near Nineteenth Street near the mixing plant. Belt
conveyors run under the gates of the bins and take the material
to a bucket elevator which raises it to bins over the mixer. The

FIG. 5. — METAL JACK ARCH FORMS.

overhead bins feed into measuring hoppers, where the cement is
added, and thence into the mixer, which is a 1-yd. machine mixing
a 5-bag batch. Conveyors and mixer are electrically driven.

Taking advantage of the firm character of the ground and
the great width of the avenue, a method of excavation was adopted
that dispensed with cross-bracing; thus leaving the cut (Fig. 6)
imobstructed and making it feasible to construct large sections
at one time. The floor of the two middle Iracks was laid firet,
and this was followed by the floor and duct bench of each of the
two outside tracks. Concrete was brought from the mixer for
this part of the work in 2-yd. side-dump cars. On the eomple-

Digitized by

Google

Noble on Use of Concrete in Brooklyn Subway. 369

tion of a section of the bottom, steel forms were moved into place,
and the sides, intermediate walls and roof were concreted con-
tinuously in a long section.

The steel forms (Fig. 7) were of the collapsible type, made
of iV"ii^' plate and stiflfened with shapes. They were composed
of 5-ft. units, bolted together in sections 40 ft. long, and traveled
on rails. Four such sections, one for each track, made up a set
of forms; of which two were in use generally on different parts

* ^

^^^^^^HS^MlBkbf 'i^9Sl^^^^3Bp^H^^V^

HKS^ jIHbis^

FIG. 6. — METHOD OF OPEN CUT EXCAVATION.

of the work. The forms were struck, each section separately,
by means of a hand tackle arrangement that swimg the top
leaves down around hinges at each side, and drew in the tops of
the side-panels automatically. The vertical reinforcement
(Fig. 8) for the next section being set up, the forms were then
pulled ahead, one 40-ft. section at a time, by a locomotive, into
its new position; an operation that could be performed in a
very few minutes. As each section was advanced it was adjusted
to exact position and connected to its neighbor by long bolts and

Digitized by

Google

370 Noble on Use of Concrete in Brooklyn Subway.

pipe separators. The adjustment was effected by the jack-
screws and sliding axles of the trucks; affording a range of move-
ment both vertically and horizontally. The reinforcing bars
of the roof were then laid in position.

For concreting the roof, the mixer discharged into trains
of 4 or more 2-yd. bottom-dump buckets on flat cars, which were
pushed by a locomotive over the tramway through the cut to the
forms. Here the buckets were lifted by a locomotive crane

FIG. 7. — COLLAPSIBLE STEEL FORMS FOR ROOF AND SIDES.

(Fig. 6), and dumped on a hopper raised about 12 ft. above the
forms. From the hopper the mixture flowed through shallow
troughs to the points required. When both sets of forms were
joined together, an 80-ft. section, requiring over 600 cu. yd.,
could be concreted in about 15 hours.

Section ll-A-4, from Twenty-seventh Street along Fourth
Avenue to Forty-third Street, is being built by the E. E. Smith
Contracting Company, who are also the contractors for section
ll-A-2.

Digitized by

Google

Noble on Use of Concrete in Brooklyn Subway. 371

Materials are delivered into bins in the contractor's yard
at Thirty-first Street and Second Avenue, where there is also a
cement storehouse of 30-cars capacity to serve both sections.

The methods of mixing and distributing concrete and the
use of forms, are the same as those described for section ll-A-2.

All cement is inspected at the mills and shipped in sealed

no. 8. — VERTICAL REINFORCEMENT.

bags. The testing laboratory is in Allentown, Pa., in the Lehigh
Valley district. Over half a million barrels are required. Under
the specifications, preference is given to brands whose records
show continued increase in strength over long periods. As a
criterion of this quality, mortar briquettes are required to show
a gain of at least 50 lb. between the 7 and 28-day tests.

Digitized by

Google

THE USE OF REINFORCED CONCRETE IN HYPO-
CHLORITE WATER PURIFICATION WORKS.

By Walter M. Cross.*

During the year 1911 an experimental installation of the
hjT)ochlorite process for the approximate sterilization of the
entire municipal water supply of Kansas City was so remarkably
successful in diminishing the sickness and death rate in the city
on account of typhoid fever as well as other forms of intestinal
disease, that the Kansas City Fire and Water Board undertook
the construction of a permanent building and apparatus for the
application of this purification process to the water supply.

A separate building was constructed to make possible the
satisfactory storage, handling and making of the solution of hypo-
chlorite ready for mixing with the sedimented water. The build-
ing itself was designed by W. C. Root, an architect, and the appa-
ratus for use in connection with the sterilization process was
installed under the direction and supervision of Burton Lowther,
engineer in charge, and S. Y. High, superintendent of the Water
Works Department.

The apparatus for the handling of the hypochlorite and the
supports for it are of reinforced concrete. It is to be observed
that no other material is so well suited for use in connection with
this sterilizing agent as good concrete for the reason that all other
materials that are capable of oxidation are promptly attacked
by the hypochlorite solution and become rapidly deteriorated.
The prime consideration with regard to this class of installation
is to employ such methods of construction and to use material
that is so permanent in character as to obviate the necessity
of repairs which would force the discontinuance of the application
of the sterilizing agent even for an hour.

The basement of the building is used fcr storage of the reagent
that is kept in reserve. The main floor is used to house the dilu-
tion tanks and the feeding devices, while on the floor above is
placed the tank in which the hypochlorite is reduced to paste

•City Chemist. KanBas City. Mo.

(372;

Digitized by

Google

Cross on Concrete in Hypochlorite Plant. 373

of a creamy consistency before being delivered to the dilution
tanks beneath. This pasting tank, 3 ft. in diameter and 4 ft.
high, is provided with a stirring device carrying two rather heavy
rollers, disposed horizontally at its lower end. The rollers clear
the bottom of the concrete tank only by a fraction of an inch,
thus insuring the mashing and disintegration of all of the small
lumps that are invariably present in commercial calcium hypo-
chlorite. Owing to the fact that the action of the reagent on
bronze is to form on the surface of it a fairly insoluble and pro-
tective coating of metallic carbonate and oxychloride, that metal
appears to be the most available for use on all bearings and stirring
or disintegrating devices that come in contact with the solution.
Leading from the concrete pasting tank are pipes so arranged
that the contents of the tank may be discharged into either of
the large dilution tanks on the floor beneath. The outlet of the
pasting tank is placed at a considerable distance above its bottom
so as to avoid the possibility of drawing off with the paste any
fragments of considerable size. The pipes carrying the paste
are so arranged as to be readily cleaned in a few minutes in the
event that they become clogged. Ultimately they are sure to
become clogged if they are not occasionally cleared because of the
formation in them of carbonate from carbon dioxide absorbed
from the air.

The dilution tanks are hexagonal in form, 9 ft. in maximum
diameter and 7 ft. high; the walls are 6 in. thick. Although
the difficulty experienced in properly disposing the reinforcing
metal in the construction of a hexagonal tank is much greater than
is the case in the building of a round one, the hexagonal tank is
to be preferred on account of the fact that in a round tank a
rotary stirrer does not produce nearly such thorough agitation
and mixing of the solution of hypochlorite as the same stirrer can
do in the hexagonal tank. The paste is mixed with water in the
dilution tanks until a uniform solution of a strength of 2 per cent
occurs. The use of the two tanks makes it possible to accurately
adjust the strength of the solution in one dilution tank while the
contents of the other are being utilized. The dilution tanks are
placed on supports high enough to permit the use of a gravity
feed to the orifice box which is placed on the floor of the room
housing the big tanks.

Digitized by

Google

374 Cross on Concrete in Hypochlorite Plant.

I I

j£^ 1

4-1-*

FIG. I.

Digitized by

Google

Cross on Concrete in Hypochlorite Plant. 375

Bronze pipes, 1^ in. in size, so arranged as to be readily cleaned
in the event of stoppage, connect the dilution tanks with a gauging
tank, 4 ft. in diameter. This gauging tank contains a float,
scale and pointer so arranged that the man in charge can accu-
rately check the speed of outflow of solution from the orifice box
into the big water main carrying the entire city water supply
from the settling basins to the pumps. The solution passes through
the gauging tank to the orifice box. Each division on the gauge
represents 1 gallon of the hypochlorite solution.

The orifice box is oblong in shape and carries a float of about
250 cu. in. displacement. The float operates a valve which,
by either opening slightly or closing, maintains the hypochlorite
solution in the orifice box to a constant level. One end of the
orifice box is of plate glass to enable the operator to see at a glance
that the solution is filling the box to the proper height. Attached
to the plate glass and covering a hole in it, is a hard rubber disc
having near its periphery several slits, the adjustment of which rep-
resents the size of a stream of the 2 per cent hypochlorite solution
that will be the proper amount to treat the quantity of water
passing through the main. All movements of the hypochlorite
solution after its preparation are by gravity. Ample opportunity
for the hypochlorite after its addition to the water to react with
any putrescible organic matter and germs, is afforded during, the
time in which the water passes through the centrifugal pumps, the
flow line and a small storage basin at Turkey Creek before it is
pumped to the domestic water users.

All of the stirring devices are run by an electric motor belted
to a line shaft carrying clutches so placed as to make possible
the running of any one of the stirrers whether or not any of the
others are running.

The principle involved in the construction of practically
all hypochlorite installations for the purification of water by the
oxidation of germs and putrescible organic matter in municipal
water supplies is substantially the same as that in Kansas City.
Concrete, usually reinforced, is universally used in the construc-
tion of all permanent apparatus for the preparation and solution
of the hypochlorite for mixing with the water to be purified.

A fairly good idea of the disposition of the various parts of
the purification installation is given in Fig. 1.

Digitized by

Google

DESIGN AND CONSTRUCTION OF THE ESTACADA

DAM.

By Hermann V. Schreiber.*

Located in the northwestern section of the country, where
fuel costs are high, it is natural that the Portland electric com-
panies should have early appreciated the abundant stream flow
characteristic of the region, which results from the high precipi-
tation on the western slopes of the Cascade Mountains and in the
Willamette Valley and which was rendered available for their
power purposes by the successful development of high tension
electric transmission for the distances involved in delivering such
hydro-electric power to their markets.

The first plant developed to serve Portland, and one of the
first water-power transmission plants on the coast, is located at
the falls of the Willamette, adjacent to Oregon City on the
Willamette River, which stream flows through Portland and
empties into the Columbia River about six miles below the center
of that city. Some years later, as the community rapidly grew
in size and industrial activity, a combination plan was projected
by interested parties for the construction of a railway and power
development on the Clackamas River, based upon an available
power site some forty miles east of Portland, and this develop-
ment has since been completed and, with the other railway and
lighting interests, has been included in a consolidated property
known as the Portland Railway, Light and Power Company,
which at present supplies practically the entire electrical market
in Portland, including the local and interurban railway systems.

Following the consolidation, the rapid increase in the power
requirements of the community indicated the need for immediate
additions to the generating capacity, and Sellers and Rippey,
Consulting Engineers of Philadelphia, were retained by the finan-
cial interests in control of the consolidated property to investi-
gate and report upon certain features of the existing plants and

* Sellers and Rippey, Consulting Engineers, Philadelphia.

(376)

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM. 377

the power possibilities and costs of certain new developments
which were suggested for consideration from time to time during
the investigations. Because of the exceptional water-power
possibilities of this region, the number of alternative locations
presented for consideration was greater than would ordinarily
come within the radius of economical transmission to a large
city, but attention was chiefly directed to the Clackamas River,
upon which the company already owned the Cazadero develop-
ment, the electric railway system and the transmission lines
connecting the generating station with Portland.

This stream has its source in the forest-covered western
slopes of the Cascades and the snow-covered peaks adjacent to
Mount Hood. The run-oflf resulting from the high precipi-
tation peculiar to this section is distributed in a remarkably
uniform stream flow which, because of the favorable geological
formation, is maintained even during the long dry summer sea-
son. Consideration was primarily given an available site for a
large new development upon property already owned by the
company on this river above the existing Cazadero plant, but
investigations extending over the thirty miles of the river down-
stream to its mouth revealed a site capable of economic develop-
ment, which, though it did not offer as large capacity as the
upper site suggested, was foimd to be worthy of recommendation
for the company's immediate consideration, because of the con-
siderably lower cost involved and sufficient size to meet the
immediate power requirements. This site, by far the most
attractive on the lower river, is a short distance below the com-
pany's town of Estacada, about 3^ miles below the original
development at Cazadero, and offered the advantages of a rea-
sonable head for direct development without long flumes, acces-
sible railway connections and several incidental advantages from
a construction standpoint. After some delay, the necessary
property was acquired and executive decision given that the
construction proceed without delay. Within a distance of a
few hundred feet several possible dam sites were available, giving
the same head, with advantages and disadvantages peculiar to
each, making the proper selection one of considerable difficulty.
Fig. 1. gives a general view of the Estacada dam, a phn of which
is shown in Fig. 2.

Digitized by

Google

378

SCHREIBER ON EsTACADA CONCRETE DaM.

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM.

379^

X
O

Digitized by

Google

380 SCHREIBER ON ESTACADA CONCRETE DaM.

It is of considerable interest at this point to consider further
the character of the country, its geologic history and bed-rock
formation as affecting the selection of site, preparation of founda-
tions and design and construction of dam.

The mountains which form the source of the water supply
include several extinct volcanic peaks, the eruptive discharge
from which has covered the country for miles around with various
forms of volcanic debris of such porous and uncertain nature that
it introduces serious obstacles to satisfactory hydro-electric
developments of any considerable magnitude, even when they are
limited to the use of a low dam and extended canal such as that
which has been installed at Cazadero. In order to properly study
the situation and secure all possible information relative to the
formation of this material which now constitutes the "bed rock"
a careful examination of this section was made on request by Mr.
J. S. Diller of the U. S. Geological Survey, who in his reports on
the subject made the following statements:

The volcanic breccia (bed rock) is made up of unassorted angular frag-
ments of lava andesite and basalt of various colors ranging in size from dust
particles and grains of sand to large rock fragments many feet in diameter.
This fragmental material was blown by explosive eruption from the volcanic
craters higher up on the range and fell upon the mountain slopes where it
became so saturated with water from the copious rains accompanying the
eruptions that it flowed in great steaming sheets from the Cascade Range to
the gentle slope of the plains, in much the same way as similar material
flowed down the old stream channels on the western slope of the Sierra
Nevada in California and covered the early and often rich deposits of aurif-
erous gravels.

Sheets of solid nonfragmental lava forming part of the bed rock and out-
cropping on the slopes of the canyon occur within and between the great
sheets of volcanic breccia. Some of the lava sheets are basalt, others are
andesite and they are usually less than 30 ft. in thickness. The basalts are
generally very porous and gray or dark. The andesites are often reddish and
porphyritic with white crystals of feldspar.

The depth to which these sheets of volcanic breccia and lava extend
cannot be readily determined but it is certainly hundreds of feet and may be,
as it is along the Santiam and McKcnzie River canyons, over a 1000 ft. in
thickness.

Nearly vertical dikes of basalt cut up through the sheets of volcanic
breccia and lava and outcrop on the surface. These dikes in some places
have a well developed columnar jointing which divides the rock into columns.
In the case of the dikes the columns lie horizontally and extend across the
dike. In the lava flows the columns are vertical, but in all cases the columnar

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM. 381

joint cracks are limited to the dike or lava sheet and do not extend into the
adjacent rock nor make an opening of great extent. There is, however,
another set of parallel joints, the open cracks of which cut up through the
volcanic breccia and sheets of lava about vertical in a direction approx-
imately parallel to the course of the canyon. Such joints may be of consid-
erable extent and form important openings for the circulation of water;
Such joints may be expected and should be carefully looked for where the
rock b covered with soil or gravel. It is especially significant that the dikes
are approximately parallel to these joint cracks and suggest that the joint
cracks may extend to great depths.

The conditions that confront the engineer along the Clackamas River
in the volcanic breccia plain region are very much the same as will be found
all along the western foot of the Cascade Range from the Colxmibia River in
Oregon to Feather River in California, one of the most import,ant water-
power belts in the United States, and the successful solution of the problem
which it presents at one point will greatly facilitate the work elsewhere.

The original development at Cazadero included a rock fill
timber crib dam covered on the up-stream slope with a large
quantity of surface soil and gravel, used to reduce to a minimum
the leakage beneath the structure. The dam creates less than
one half the operating head, the balance resulting from the exten-
sion of a head race canal and flume about If miles long to the
generating station, the tail water of which is much lower than
the base of the dam because of the intervening slope of the river.
This represents a conmion method of development in the West^
where first cost, rather than low operating, maintenance and
depreciation charges, has in the past largely governed such pro-
jects. This development was well advanced when taken over by
the present owners and as completed by them utilized conditions
advantageously, although it possesses in a degree the inherent
disadvantages of such construction in that its peak load capacity
is limited by the forebay pond capacity and the leakage under the
dam and from the canal tends to reduce the dry season plant
capacity. The present owners have however succeeded in reduc-
ing the leakage under the dam to a trifling amount.

Considering the magnitude and permanence- of the consol-
idated property, it was felt desirable in planning extensions to
endeavor to provide for direct power developments, creating the
entire head by a concrete dam affording large storage reservoir
capacity and locating the plant directly at the dam without inter-
vening flume or canal, thus permitting almost indefinite peak

Digitized by

Google

382 SCHREIBER ON ESTACADA CONCRETE DaM.

Joad capacity to be developed. This method of development
involves higher dams and consequently demands greater assur-
ance concerning the security of foundations and before asking
Mr. Diller to report upon the geological formation core drill
investigations were started to determine the characteristics of
the underlying material. His report, as will be noted, strongly
confirmed our own conclusions as to the importance of this prob-
lem and we continued our extensive explorations with core drills,
going as deep as 250 ft. at times, and not content with a partial
examination, we made investigations at considerable cost and
trouble in the river bed as well as upon both banks.

SITE.

At the first new site considered for the development of power
•on the Clackamas River above Cazadero it was proposed to
create 135 ft. head at a direct connected plant, providing large
pond storage which would be of considerable value to all the
plants existing or constructed at any future time on the river
below this point.

At the Estacada site below Cazadero it was proposed to
develop power under 83 ft. head, utilizing all the available fall in
the river in the 3J miles below the original Cazadero power sta-
tion and providing pond storage which would be of great value
in carrying the daily peak load.

The investigations for the higher head development at the
upper site were started before the lower property was acquired
and were under the direct supervision of the field engineer, Mr.
Shirley C. Hulse, under our direction. The very thorough study
of the situation there gave considerable data for use in develop-
ing a method of treatment to insure as far as it might be prac-
ticable an absolute cut-oflf across the canyon which would prevent
any undue leakage or erosion of the river bottom after completion
of construction, and the results of this work were immediately
applied to the Estacada construction, saving much time in the
preliminary engineering there.

When the company was ready to proceed with this construc-
tion the work was very urgent. The general preliminary inves-
tigations covered both sides of the river for a distance of several
hundred feet; well drillings were made without finding any

Digitized by

Google

SCHREIBER ON EsTACADA CONCRETE DaM. 383

radical difiference in the character of the bed rock, except that
the first and the last of the several sites considered were found to
be underlaid with a soft clay formation which might have neces-
sitated a reinforced mat or piling to support part of the struc-
tures. Of the four sites here considered, the third one was selected
as offering an island which it was expected would considerably
reduce the amoimt of material required and at the same time
make the construction much more convenient by facilitating the
diversion of the stream through the rainy season.

Further investigations were then made of the actual condi-
tions on the island with the result that a large ravine was foimd
to extend parallel to the thread of the stream, which being filled
with debris necessitated its entire excavation. As the construc-
tion proceeded the left end of the island adjoining this ravine was
found to contain a layer of clay sloped in such a way as to endan-
ger its stability if loaded with any of the dam superstructure.
This was also removed and with the removal of another section
of the island on account of the great uncertainty of the effective-
ness of any device against leakage there was little left of the
island and no expense was saved by its use in the construction.

CUT-OFF WALL.

The study of the foundation conditions on the Clackamas
River was evidenced by the investigations and test pits, together
with rough pressure tests, Mr. Diller's expressions relative to
the geological formation, and the general hesitancy among engi-
neers with respect to constructing high masonry dams upon such
foimdations confirmed the original belief that somewhat novel
methods must be adopted to give reasonable assurance concern-
ing the integrity of the work to be constructed. Moreover, it
was evident that the nature of these methods should be such as
to permit demonstration of their efficacy before any large invest-
ment in the construction should be made, and this involved some
experimental work and expense which would not be involved on
more substantial or satisfactory foundations.

It will be evident that impermeable foimdations are desirable
for the following reasons:

(a) To minimize the possibility of upward pressure under the

Digitized by

Google

384 SCHREIBER ON EsTACADA CONCRETE DaM.

base of the dam superstructure. (This applies chiefly to solid
dams.)

(b) To prevent percolation under the dam which might lead
to sufficient erosion to involve undermining of the structure.

(c) To overcome any structured weakn^ess due to the original
geological formation and properly provide a support for the
superimposed load.

(d) To avoid waste of water (the chief power asset of the
company) from the reservoir, through, under or around the dam,
instead of through the turbines, thus providing K.W.H. for sale.

To satisfy these conditions with existing foundations of the
character described in Mr. Diller*s reports required a departure
from any methods heretofore used of which we had knowledge
and in studjdng the problem it appeared that the only safe pro-
gram must practically provide for actually changing the struc-
tural character of the underlying formation. It occurred to our
chief engineer, Mr. S. Howard Rippey, who had spent some time
on the ground, that the most promising method would be the
solidification of the porous material by the introduction of cement
grout under pressure, but that the success of the work must be
susceptible of demonstration by actual test before the super-
structures should be started. Thorough inquiry failed to disclose
any precedent for the use of grout for the general treatment of
foundations, although it was foimd that cavities in limestone rock
under the New Croton Dam had been filled with grout, much as a
dentist would fill a cavity in a tooth. The grouting method had
also been used in filling back of lining walls in tunnels, etc.
Notwithstanding the absence of precedent, it was decided to pro-
ceed with a grouting scheme and a program was outlined for
preventing leakage of water from the reservoir created by the
dam, under or around the dam, to the low tail water level below
the dam and the experiments which should be made to properly
demonstrate the efficacy of the method before the complete
development should be imdertaken were prescribed. The general
idea provided for drilling a double line of holes of an average
depth of say 50 ft. under the heel of the dam across the entire
valley to and under the shore abutments and the subsequent
forcing into each of these holes of grout of such consistency as to
percolate through the entire substructure and so permeate it as

Digitized by

Google

SCHREIBER ON EsTACADA CONCRETE DaM. 385

to solidify it absolutely throughout the entire length of the super-
structure, thus making the foundations absolutely solid and the
equivalent of a deep cut-off wall.

It was recognized that experiments would be necessary to
determine the proper spacing of the grout holes and their depth,
which would ensiu^e sufficient diffusion of the grout through the
varying material encoimtered to create a continuous impermeable
barrier, thus preventing seepage of water from the reservoir imder
the hydrostatic head which would be created by the dam.

After drilling the double line of holes, the program con-
templated the test of each hole with water pressiu^e, a record
being kept of the quantity escaping and the pressure applied to
each hole. This water pressure test also provided for washing
out the interstices ready for the reception of the grout. Upon
the completion of the tests, the cement grout was to be pumped
into the holes imder pressure and after allowing time for harden-
ing, a third line of holes was to be drilled midway between the
first two or outer lines and tested with water pressure. The idea
was that if water pressiu^e be applied to the center holes at or
slightly above the hydrostatic pressure to which this rock would
be subjected by the water in the reservoir after completion of the
dam and no appreciable leakage occiured, we should feel rea-
sonably certain that the cement grout had proven effective in
making the entire foimdation impermeable.

Before purchasing apparatus to handle the thin cement
grout communication with several pump manufactiu*ers showed
that in so far as the manufacturers' guarantees were concerned,
choice was limited to two standard makes of hand-operated
diaphragm pumps which would require 8 men on the handles to
develop a pressure of 100 lb. per sq. in. Only one power dia-
phragm pimip was offered by the manufacturers who would
assume no responsibility for its successful operation. It was
foimd that in both this and foreign coimtries plimger, diaphragm
and centrifugal pumps had been successfully used, but after care-
ful study of all the commercial pumps available it was decided
that the use of compressed air would be more effective, flexible
and economical than any mechanical pump. Having eliminated
the question of pumps it was foimd that several different types
of compressed air tanks were available; some being provided with
different numbers and shapes of blades which were revolved in

Digitized by

Google

386

SCHREIBER ON EsTACADA CONCRETE DaM.

the tank for the mechanical mixing of the cement grout while in
others the grout was mixed entirely by the circulation of air.

Interviews were had with a number of contractors and engi-
neers who had had occasion to use the different types of machin-
ery and it was finally decided to purchase the Canniff Pneumatic
Grout Mixer and Injector. This machine, Fig. 3, consists of a
plate steel cylinder with a conical shaped bottom and flat plate

Blow off when nixing

FIG. 3. — CANNIFF PNEUMATIC GROUT MIXER AND INJECTOR.

steel top which is provided with a smaller hinged circular lid
opening inward through which the water, cement and sand are
introduced. The tank is provided with a blow-off valve on top
of the lid, also an air inlet pipe tapped into the side near the top,
a grout discharge pipe tapped into the bottom and a by-pass pipe
between the air and grout pipes. The valves are so arranged that
when the cement, sand and water are placed in the tank the com<<

Digitized by

Google

SCHREIBER ON EsTACADA CONCRETE DaM. 387

pressed air is forced into the bottom of the tank through the
grout pipe and having free escape through the blow-off valve at
the top, the grout is thoroughly mixed by the passage of air
bubbles through it.

When it is desired to discharge the grout the valves are
so manipulated that the air is introduced near the top of the
tank and the grout forced out through the bottom pipe, the press-
ure being limited only by the capacity of the compressor and
the strength of the tank. It is possible to connect any number
of these tanks on to one main grout discharge pipe, thereby insur-
ing a continuous flow of grout and by a proper system of pipe
connections any one tank may be cut out for repairs without
interfering with the operation of the remaining tanks.

After experimenting on the Estacada foundations with both
the shot and diamond core drills, an equipment of 7 Calyx shot
drills was finally installed and operated at this development
night and day for a period of about one year. These drills gave
a hole varjdng from 2f to 8 in. in diameter depending upon the
character of the bed-rock material.

Where possible, in advance of the core drilling 3-in. wrought
iron pipe casings were set to a depth of from 4 to 6 ft. in the rock
and grouted, the upper ends projecting about 1 ft. above the
surface and being threaded to permit the attachment of the
testing and grouting apparatus. While the original plan called
for grouting in advance of any construction, the failure to start
this portion of the work at once tended to retard the actual con-
struction work and it therefore became necessary to construct a
concrete cut-off wall 7 to 10 ft. deep and set the casings in this
wall and drill the holes through the casings into the rock below
where this was necessary. The cut-off wall was placed upstream
from the foot of the dam and the construction of the dam super-
structure was continued without interference.

The final report of Mr. Frank R. Fisher, resident engineer,
for the Light and Power Department, Portland Railway, Light
and Power Company, imder whose supervision the Estacada con-
struction work was completed, gives the following additional
detail information relative to the conditions which existed and
methods used in the grouting:

The rock mass is traced throughout with Beams, very irregular in shape
and size, extending in all directions with but slight continuity. They vary

Digitized by

Google

388 SCHREIBER ON EsTACADA CONCRETE DaM.

from an almost imperceptible cleavage joint up to those having an approx-
imate width of from one to two inches, the large majority observed meas-
uring but a fraction of an inch. The seams are more or less choked with
sand, gravel, small particles of rock, and other debris. No large crevices or
faults were found at the site, nor observed in the vicinity of the dam.

On the whole, the rock gives indication of possessing fair bearing value,
and while not what would be classed as hard, would probably offer consid-
erable resistance to the erosive action of water, except under high velocities.

The plan was adopted of distributing drills over a wide area and each
hole drilled was tested and grouted immediately on completion before putting
down any other nearby holes. In grouting, the method usually followed was
to make connection between the grout tanks and the casing by means of a
flexible copper hose, and introduce the grout at the top of the hole; but in
order to prevent, if possible, the rapid choking of the hole with cement, which
frequently occurred, the method of introducing a pipe into the hole, and dis-
charging the grout at various depths was tried out. For this purpose a 2-in.
pipe, made up in sections, was used, and the operation was started with the
same inserted to within a few feet of the bottom of the hole. When the hole
gave evidence of tightening with the pipe in this position, one section was
detached, usually 10 ft. in length, thus raising the outlet, and the operation
repeated. At intervals a charge of water was shot in, to keep the pipe from
plugging, and also to loosen up the cement that settled in the hole. This
method of grouting through the pipes was given a thorough trial, but so far
as could be observed it had very little advantage as to the amount of grout
the hole would take, over the less laborious operation of introducing it
directly at the top. The changing of the position of the pipes also interrupted
the continuous flow of the grout, which it was desirable to maintain in order
to accomplish the best results.

The consistency of the grout was varied to meet the different conditions,
1 part of cement to 5 of water appearing to give the best results, but the
proportions tried out varied from 1 cement and 2 water to 1 cement and
15 water.

The grout was forced in under air pressure ranging from 50 to 200 lb.
per sq. in., depending on the tightness of the hole, but the material at the
lower depths would not tighten up beyond a certain degree. While it was
desirable to use the higher pressures, in order to accomplish the greatest
diffusion, it was not always practicable; to do so, on account of it blowing
out at the surface.

As the general nature of the rock had been investigated through the
cores from the preliminary exploration holes, no special effort was made to
preserve the same when drilling the primary holes. That from the proving
holes, however, was carefully examined for any traces of grout, and while
some very fair specimens of cement core were obtained, the amount was not
large. This was probably due to the fact that most of the seams were small,
and the cement entering therein would be ground to powder under the action
of the bit. It is also probable that in many instances, at the time the prov-
ing holes were put down, the cement had not set sufficiently hard to core.

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM. 389

as the conditions arising from the various stages of the construction work,
as well as the rate of progress to be maintained, made it necessary to follow
on with the proving, within a short time after the grouting of the primary
holes.

The pressure testing of the proving holes was important and the final
plan adopted gave a direct pressure from tanks located above the proposed
pond water level as best approaching the final conditions. After the piping
was filled, a test run of ten minutes was taken, the rate of seepage per minute
being averaged and recorded. Besides testing individual holes, severs!
combinations were tested in order to determine if possible the extent of inter-
communication beneath the surface. The combined seepage in this was
found to be considerably less than the sum of the separate tests and in some
instances was only about half. An additional refinement was introduced by
the testing of each proving hole in some sections every 10 ft. in depth as it
was being treated, thereby giving an indication of the effectiveness of the
grouted cut-off wall at different depths. In some instances communication
was found to exist between holes located as far as 70 ft. apart.

After the completion of the dam, and subsequent filling of the pond, two
test holes were put down inside of the dam, approximately 30 ft. from the
cut-off. One of these was located between buttresses Nos. 10 and 11 and the
other between Nos. 12 and 13, opposite the weakest points in the grouted
cut-off. After drilling to a certain depth, backfiow was obtained from each,
and the maximum height to which the water rose in an extended pipe, due to
upward pressure, was within 13 ft. of the elevation of that of the pond. The
results of these tests indicate that some upward pressure exists in the founda-
tion material in this locality, which the mass of overlying rock is resisting.

The excavation of a supplementary cut-off trench at one point where
grouting had previously been done offered excellent opportunity for observing
its effects, as many seams were exposed and all of them proved to be well
caulked with cement.

The data of quantities and cost of this portion of the work is also of
interest.

Drilling 555 Holes, 34,038 Pr., Costino $1.55 per Pt. for Drilling

AND Grouting.
Labor —

DrilUng $0.68

Grouting 18

Repairs 17

Plant (drills and grout tanks) 30

Cement 12

Power 05

Overhead (incorporated general plant) 32

Total $1.72

Less salvage 17

$1.55

Digitized by

Google

390 SCHREIBER ON EsTACADA CONCRETE DaM.

In one section of the work the original pressure tests on the
ungrouted holes frequently developed a seepage of over 100
gallons per minute per hole and communication existed between
holes located over a wide area and surface leakage also appeared
in many places. Where the original primary holes gave an
average leakage before grouting of 80 gallons per minute, thirteen
proving holes when drilled showed an average leakage of 7 gallons
per minute. Another section, for the first 30 ft. of depth below
the concrete cut-ofif the treatment appeared to be fairly success-
ful, but at the full depth a test on a group of thirteen proving
holes showed an average seepage of 6.4 gallons per minute. In
still other sections the proving holes showed seepage varying
from 2.2 gallons to 3.6 gallons per minute.

A total of 1942 bbls. of cement were used in grouting. The
average depth drilled per hour was 1.32 ft., including time for
moving, etc. The cement pumped into the holes varied from
3.32 bbls. to 50 bbls.

This treatment indicated the possibility of constructing what
may be termed an effective cut-off without being literally imper-
vious. There is however no assurance that this degree of success
would result from similar work on another Qite with this kind
of foundation.

DESIGN OF DAM.

Previous experience in the construction of hollow dams
together with numerous studies of comparative costs and advan-
tages inherent in hollow and solid dams prompted us to recom-
mend the hollow dam for use in this instance particularly because
of the character of the foundations. WhOe several other forms
of hollow dams are available and have been carefully studied and
some have been built, the distance at which the work had to be
directed together with the urgency of the construction strongly
favored the adoption of the reinforced deck and buttress form of
hollow dam, on account of the existence of an experienced con-
struction organization which could be started at once on the work
and push it to completion without delay.

The materials and stresses specified for use in this design
were as follows:

Concrete —

Deck and apron 1:2:4

Buttresses 1:3:6

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM. 391

Reinforcement — Corrugated square bars 50,000 lb. elastic limit and 80,000
to 100,000 lb. ultimate strength.

Stresses —

Modulus of elasticity of concrete 1,500,000 lb.

Modulus of elasticity of steel 30,000,000 "

Compressive stress in concrete 500 lb. per sq. in.

Shear in concrete 75 " "

Tension in concrete " "

Tension in reinforcement 15,000 " "

Base pressures 100 " "

The hollow dam of the design selected consists of a series of
parallel walls or buttresses running parallel to the thread of the

FIQ. 4. — SECTION OP DAM THROUGH SPILLWAY.

stream, with an upstream covering or deck on one side and, on
the spillway section, a downstream covering or apron which ter-
minates at its base in a heavy curved bucket section designed to
divert and discharge the falling water downstream parallel to the
river bed. A section of the dam through the spillway is given in
Fig. 4, and through the intake in Fig. 5.

The deck, which is inclined 45 deg. with the vertical, has a
thickness that varies from 20 to 48 in. depending upon the depth.
The apron which makes an angle of 30 deg. with the vertical also
varies in thickness, being thickest at the crest and bucket but
reduced to a thickness of 18 in. on the straight slope section.
The reinforcement in deck and apron is laid horizontally 2 in.

Digitized by

Google

392

SCHREIBER ON EbTACADA CONCRETE DaM.

from the under side of the slab with vertical bars on 24 in. spac-
ing as an extra precaution against weakness in joints and to avoid
temperature cracks. Hydrated lime, 30 lb. per cu. yd., was added
to the deck, crest and apron to make the material less pervious.
The extra cost for this material averaged 23 cents per yd. The
buttresses spaced on 18 ft. centers are built in horizontal lifts of
12 ft. and vary in thickness from 15 in. at the top to 38 in. at the
bottom, being also tapered along their length for equalization of
pressure. Haunches with 18-in. seats for the deck slabs are pro-

FIG. 5. — SECTION OF DAM THROUGH INTAKE.

vided at the upstream end and tongues extended between the
slabs support the deck forms during construction and provide an
opportunity to get a tight joint between the slab and the buttress.
Because of the uncertain character of the foimdations the foot-
ings were spread to reduce the base pressure to 100 lb. per sq. in.
Additional reinforcement was also provided in the bottom lift
to permit bridging any spaces of uncertain bearing value. The
spillway section of the dam has a maximum height of about 86 ft.
and the bulkhead section a maximum height of about 101 ft.

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM.

393

As constructed at Estacada the usual features were embodied
in the spillway construction with the addition of sluice gates and
the usual closing device in the left or main river channel. The
spillway end abutment was of solid retaining wall section, a large
part of it being built in gravel and clay, serving principally as a
cut-off or core wall to prevent wash around the end of the dam.
The right-hand or bulkhead end abutment was considerably
longer, of reinforced retaining wall design.

The island which, as stated, was included in the development

iL^jUt&W.J

Fia. 6. — METHOD OF CONSTRUCTION AND FORM WORK.

was found to be of such poor material that about one-third of it
was entirely removed to better bottom about 10 ft. above the
water surface and the balance was notched out parallel to the
deck and faced with 2 ft. of concrete which formed an extension
of the deck and tied in with the cut-off wall at the base and a
low section of dam on the top of this portion of the island.

The total length of the spillway or overflow section extending
from the left abutment to the non-overflow section on the island
is 404 ft. 10 in. and has a rounded crest and heavy bucket and

Digitized by

Google

394

SCHREIBER ON ESTACADA CONCRETE DaM.

apron extending over the downstream side of the island far
enough to fully protect this portion of the structure from imder-
mining. The buttresses in this portion are pierced near the top
with openings used for a runway extending entirely through the
dam from the power house up through the left abutment.

The non-overflow or bulkhead section in which the but-
tresses extend 15 ft. higher and are covered only on the upstream
side and top, is open on the downstream side. One bay in this

FIG. 7. — METHOD OF CONSTRUCTION AND FORM WORK.

section contains a large concrete fishway tank discharging into a
concrete fishway.

Figs. 6 and 7 show the dam in course of construction and the
various types of forms are illustrated in Figs. 8, 9 and 10.

Because of the necessity of providing adequate trash rack
area, supporting penstocks and meeting the requirements imposed
by the high tension transformers and wiring installed here, the
power-house section of the dam is quite complicated and required
considerable thought and skill to work out the situation satisfac-
torily.

Digitized by

Google

SCHREIBEB ON EsTACADA CONCRETE DaM.

395

To locate the penstocks at convenient distances and properly
support them on the adjoining buttresses as they pass from the
intakes through the dam to the turbines, the spacing between
buttresses is here made 14 ft. instead of 18 ft. as in the balance
of the construction. An extra thickness of deck with additional
reinforcement is provided aroimd the penstock intakes and to
carry the weight and prevent vibration several extra heavy
struts are built against each penstock so as to tie across to the
adjoining buttresses.

To provide means for getting machinery into the power
house, which is built against the downstream ends of the but-
tresses of the power-house section, a railway connection is

«1

I
I

FIG. 8. — ^METHOD OF HOLDING TONGUE AND HAUNCH FORMS IN PLACE.

extended along the top of the right end abutment so as to enter
directly on the unloading platform where crane service is avail-
able to quickly lower material to the power-house floor level and
where, after being transferred by a suitable truck, it is handled
by the power-house crane.

The deck for the intake section of the dam is set back from
the face of the deck below it to allow space behind the trash
racks for head gates and stop logs and to permit easy access of
the water from racks to penstock opening and is of special con-
struction on this account.

The racks, hoists and loading platform are housed by a con-
crete structure extending the full length of the power house. The

Digitized by

Google

396

SCHREIBER ON EsTACADA CONCRETE DaM.

power house has a solid concrete substructure with flood protec-
tion for a height of 14 ft. above which is a concrete and steel
superstructure.

The initial installation consists of three 6000 h.p. 240 r.j).m.
twin flumC; center discharge, volute casing turbines, of special
design originally suggested by us, each connected to a 3300 k.w.,
3-phase, 10,000-volt generator, and penstock and power-house
space is provided for fwo. more units.

These turbines feed from below through a cast iron Y pipe
connection to their penstocks, have substantial cast iron bed
plates and casings, presenting a neat and compact appearance.

--^ _ / _ .-----^ _ _ Iv^'^

a^ *. io» ftevin

FIG. 9. — METHOD OF ANCHORING DECK SLAB FORMS.

The oil type governors are provided with the usual distant control
and other auxiliaries as well as extra flyball head for emergency
"shut downs" in case the regular head fails to act.

The generators are so supported on the foimdations as to
have ample ventilation in all parts. A direct connected exciter
on the outboard end of each shaft and one reserve combined
turbine and motor-driven exciter provide the necessary exciting
current.

The 3-phase, 60-cycle, 33,000-volt delta and 57,000-volt Y
step-up transformers, each carrying the load of the adjacent
generator, are located in the dam back of the turbines and inmae-

Digitized by

Google

SCHREIBER ON ESTACADA CONCRETE DaM.

397

diately above them are the 57,000-volt high tension and the
10,000-volt low tension oil switches, bus bars, electrolytic light-
ning arresters, etc., connecting with the two 3-phafie transmission
lines to Portland.

CONSTRUCTION.

The scarcity of suitable construction materials was at fii^t a
matter of considerable importance but with the assistance of
Mr. Robert S. Edwards, consulting chemical engineer, of Port-
land, who was retained to investigate the quality and costs of

BoUt to boia

flMM dott*« an MM to np-

•ort ToacM roiw oBtll boia

(7 Belt*. BmovM UUt* ft»apr«t«iv

■ttHhtiw teok Strei '
*• » 4- « 4- VMtar

1- ■ I" h»U ta CtnanU ter
4ptk-4, iB« VMf to ioppoit

FIG. 10. — FORMS FOR BUTTRESS.

sand, stone and cement available, satisfactory materials were
obtained.

Sand for construction in this section is ordinarily pumped
from the Columbia River, but on investigation of the possibilities
along the railway right of way a large pit was discovered which
provided suflScient sand and gravel of good quality at consid-
erable saving for not only this job but for much of the other
construction work which the company had on hand.

The only rock quarry opened up was also along the railway
right of way, and though far from satisfactory produced a good
quality of basalt rock at reasonable cost.

Digitized by

Google

398 SCHREIBER ON ESTACADA CONCRETE DaM.

Because of the fact that the cement manufacturing business
on the Pacific Coast was still in a rather non-«ystematized state
both as regards the standard of uniform quality and the matter
of regular supply, the purchase of cement was considered only
after a careful personal investigation and report on the available
mills on the coast to determine the history and quality of the
product and the probability of securing satisfactory shipments.
Considerable time and expense were devoted to this investigation,
but with these data in hand an advantageous contract based on
satisfactory specifications was closed for sufi^cient cement to meet
the company's requirements for several years at a considerable
saving.

The time for construction being quite limited with a large
amoimt of excavation to be made, the construction plant invest-
ment was considerable. The equipment for excavation consisted
of 8 steam drills assisted by 5 derricks, 1 elipctric locomotive and a
steam locomotive crane. For the core drilling and grouting there
were required 6 Davis-Calyx core drills, pressure tanks for water
testing and 2 motor-driven Peerless air compressors and 2 Canniff
air-stirring tanks for grout supply.

The sand pit was equipped with screen for eliminating the
large stone and coarse gravel and with several bins which were
arranged to empty directly into the railway cars which passed
beneath them. The quarry equipment included a motor-driven
crusher and screens together with bins for dumping directly into
the cars. At the site liberal sand and stone bins were provided
for each of the 2 electrically driven 1-yd. mixers first installed.
A third steam-driven mixer was later installed and the three
together were utilized to supply the two cableways which carried
the concrete directly onto the work.

Cement and lime storage sheds were also provided close to
the mixers to meet the ordinary needs between receipt of ship-
ments.

The concrete mixing plant and material storage bins and
sheds were located beyond the right end of the dam where they
were fed by electric railway supply trains from the company's
line to Cazadero. The mixed concrete was then fed to the cable-
way buckets for distribution over the work. On the buttress
forms tracks were arranged to carry a suitable concrete car which

Digitized by

Google

SCHREIBEB ON ESTACADA CONCRETE DaM. 399

received the concrete from the cableway bucket and distributed
it as required in the buttress.

The construction necessarily proceeded along lines dictated
to a considerable extent by the large excavation necessary on the
island and the time required for installation of machinery in the
power house.

Starting on the left channel during the beginning of the dry
season the cut-off wall was installed, holes drilled and grouted
and the first lift of a buttress constructed in midstream to serve
in dividing this channel during the following dry season when
completing the construction. During this time excavation and
grouting were proceeding on the left bank and island above water
line to permit concreting at these points to follow as soon as
possible. Next the right or power-house channel was per-
manently unwatered and the old river chasm found below the
bed of the stream was cleaned out and work was pushed, on grout-
ing and concreting to permit installation of racks, penstocks and
machinery in advance of the completion of the spillway. The
great amoimt of excavation on the island hindered this portion of
the work to some extent, but fortunately no unusual floods were
experienced and the entire structure up to the left river channel
was completed early enough so that with the installation of two
special openings in addition to the four sluice gates the stream
flow could be passed imtil such time as it might be possible to
discharge it over the crest.

The construction work started in June, 1910, and the water
first passed over the spillway November 7, 1911, which in view
of the conditions to be met may well be considered quite rapid
construction. The lar£:?st average force employed for any month
was 655 men. The record excavation was 9476 cu. yd. per
month. The greatest yardage of concrete placed any month was
8325 cu. yds.

Credit is due Mr. Robert S. Edwards, who had charge of the
cement investigations and testing, and Mr. Frank R. Fisher,
resident engineer on the later part of the construction, for data
used in this paper.

Digitized by

Google

UNIT COSTS OF REINFORCED CONCRETE
FOR INDUSTRIAL BUILDINGS.

By Chester S. Allen.*

Unit costs are harmless when used with judgment and
prudence, but likely to bring remorse and anguish when employed
promiscuously. Rare and talented indeed is the man who pos-
sesses the experience, judgment and intuitive sense to know
when, where and how to properly modify any tables or state-
ments of unit costs to meet the peculiar conditions of each indi-
vidual case. While the figures given in this paper are all taken
from structures erected during the past two years under the
writer's supervision, the wide range of territory, local condi-
tions, and different seasons of the year imder which the various
pieces of work have been executed are so great as to render the
information of value only in a very general way.

As a general proposition it has been found that reinforced
concrete is the lowest-priced fireproof material suitable for fac-
tory construction and while it is true that its first cost will gen-
erally run from 5 -to 20 per cent higher than first-class mill con-
struction, recently in several instances, with lumber at a high
price, reinforced concrete has worked out cheaper than brick and
timber. It is especially adapted to heavy construction and for
heavy loads of 200 lb. per sq. ft. and over where the spans are
18 to 20 ft. centers, not even timber can compete with it.

The unit costs of projected or completed buildings are com-
monly figured either as so much per cubic foot or as so much
per square foot of area occupied. Table I gives the unit costs
both on the sq. ft. and the cu. ft. basis, together with a general
description of a number of reinforced concrete industrial buildings
of different types erected during the past two years. It will be
seen from an examination of this table that the average cost per
sq. ft. of these buildings, excluding the one-story structures, was
$1.12; while the average cost per cu. ft. was 8.7 cts. The one-
story structures both had reinforced concrete sawtooth roofs

*■ Engineer, Lockwood, Greene & Company, Boston, Mass.

(400)

Digitized by

Google

Allen on Unit Costs op Reinforced Concrete. 401

<
i

o
O

o
U

I !
fl

HqqM^MM^MM

-33

g gcQQQ'O'aQQGOCQ no 00 53

S«^3SS||3 S 3 3

|S2SS§|§

lOlQ

^ 09 lO 00

^e^^ioe^e^^^^ o ^ e«

£3

II
.lgb|.||||i§i§ Sm

j $ S Q 5

3 .

QQ OQ

Digitized by

Google

402 Allen on Unit Costs of Reinforced Concrete.

and the average cost per sq. ft. was $1.77, while 8.5 cts. was the
average cost per cu. ft. The above costs are for the finished
buildings, including plumbing, but do not embody heating, light-
ing, elevators, sprinklers and power equipment. The cost per
sq. ft. of floor area was obtained by dividing the cost of the
building by the total number of sq. ft. of floor area exclusive
of roof area but including basement floors; and the cost per cu.
ft., by dividing the cubical contents into the cost of the structxire.

While no coal pockets are included in Table I, it has been
our experience that above 3000 tons capacity reinforced concrete
elevator coal pockets cost from $5.50 to $7.50 per ton of capacity.
Standpipes, exclusive of the foundations, average from 2^ to 3 cts.
per gallon of capacity.

On much of the reinforced concrete work which has been
done imder our supervision it has been possible, owing to the
contract being either on a percentage or cost plus a fixed sum
basis, to obtain quite accurate and comprehensive cost data.
This data, of course, is only of particular value when all the
local color of each specific case is known, but the average results
are at least interesting.

The average unit cost of the 1-2-4 concrete in the floors
including the beams, girders and slabs, was $6.10 per cu. >d.,
and for the columns $6.70 per cu. yd. Where 1-1^-3 mixture
was used for the columns the average cost was $7.60 per cu. yd.
This cost was made up of the items of cement, sand, stone or
gravel, labor and plant. The cement of course varied greatly
with the demand, but the average net cost was $1.35 per barrel
including 3 cts. for tests. The sand averaged 80 cts. per
cu. yd. and the crushed stone $1.25 per cu. yd. The cost of
labor of unloading the materials and mixing and placing the
concrete varied from 65 cts. to $2.90 per cu. yd. The cost of
plant, consisting of freight, depreciation or rental of mixing and
hoisting towers, erection of same, power and coal, and losses
and waste on the small tools, ranged from 50 cts. to $1.50 per
cu. yd. of concrete placed.

Next to the proper design of the structural features of a
concrete building, the economical design of the form work is of
paramoimt importance. The truth of this statement is borne
out by the fact that on the average job the cost of the forms

Digitized by

Google

Allen on Unit Costs of Reinforced Concrete. 403

amounts to about one-third the cost of the entire structure. On
the buildings under consideration the average cost of the forms
for the floors, including beams, girders and slabs, was 10 cts.
per sq. ft., and for the columns 13 cts. per sq. ft. The lowest
cost was in a building of the girderless or flat slab type of con-
struction, where by the intelligent use of corrugated iron for
the slab forms the cost of the floor forms, including wall beams,
was 7 cts. per sq. ft. The highest cost was for an artistic but
not elaborate overhanging cornice on a 12-story building, and
was 32 cts. per sq. ft. This last item rather forcibly demon-
strates that any attempt at architectural development is very
apt to be a costly proposition.

The cost of the labor of making, erecting and stripping the
forms varied according to the price of lumber, design of the
structure, method of forming, character of the supervision and
the skill of the workmen from 4J to 12 cts. per sq. ft. The cost
of lumber, nails and oil divided by the sq. ft. of forms
averaged from 2J to ^ cts. per sq. ft.

The cost of bending and placing the reinforcing metal, in-
cluding the necessary wire, averages $10 per ton, the range being
from $5.75 to $17.20 per ton.

Granolithic floor finish l}-in. thick when laid before the
concrete below it had set so as to form one homogeneous slab,
cost on the average of 4^ cts. per sq. ft. When put on after the
rough concrete slab, the cost averaged 7 cts. per sq. ft.

Inasmuch as the only economical design of a reinforced
concrete structure is one which closely resembles that of the
steel skeleton type, the relative cost of the various materials
commonly used for curtain walls under the windows may be of
interest. The writer has used brick, vitrified tile, concrete blocks,
cast concrete slabs and solid concrete walls for this purpose.

The most common type of curtain wall has been either an
8-in. or 12-in. brick wall resting on the concrete wall beam. The
average cost of these walls has been 45 cts. per sq. ft. There
is practically no difference in cost between the 8-in. and the
12-in. brick curtain wall, as the saving in material is offset by
the great amount of extra labor in culling and laying the thinner
wall.

An excellent and inexpensive curtain wall is constructed

Digitized by

Google

404 All^n on Unit Costs op Reinforced Concrete.

by using 8 x 12 x 18 in. vitrified tile. This is a non-absorbent
wall and when properly laid in cement mortar makes a tight
weather-proof curtain wall. The cost of this wall averages about
25 cts. per sq. ft. If the tile is plastered both sides, the cost is
about 38 cts. per sq. ft.

Where 8-in. concrete curtain walls were cast in place after
the skeleton frame was completed, the average cost was 40 cts.
per sq. ft., and when poured simultaneously with the columns
48 cts. per sq. ft. 4-in. cast concrete slabs cost about 35 cts.
per sq. ft.

While concrete blocks make a very cheap and light curtain
wall, the price being about the same as for the 8-in. tile, the
writer's experience with them has been rather unfortunate on
account of the extreme porosity of the blocks used.

Where the location of the buildings has demanded special
treatment of the exposed surfaces, they have generally been
specified to be rubbed with a block of carborundum. The aver-
age cost of this work has been 4 cts. per sq. ft. In two instances
portions of the structures have been bush hammered with a result-
ing average cost of 7 cts. per sq. ft.

Concrete piles were used on the foundations of several of
the buildings and the average cost of the piles was $1.15 per
lin. ft.

The most common methods of waterproofing concrete struc-
tures are by the introduction of foreign ingredients into the con-
crete, by the application of a compound to the concrete surface,
by the use of paper or felt waterproofing, and by accurately
grading and proportioning the aggregates and the cement.

Where an addition of hydrated lime in the proportions of
10 per cent to the weight of the cement has been used, the added
cost to a cubic yard of 1-2-4 concrete has been 50 cts. Patented
compoimds have cost from 25 to 35 cts. per sq. ft. of surface
covered. On horizontal or inclined surfaces, we have sometimes
used a granolithic surface of rich mortar of Portland cement and
sand or Portland cement and screenings in the proportions of
1-1, laid at the same time as the base and troweled as in side-
walk construction. The cost of this work has been about 5 cts.
per sq. ft.

Taken as a whole, the lowest possible cost on a reinforced

Digitized by

Google

Allen on Unit Costs of Reinforced Concrete. 405

concrete building can be obtained only by a careful study of
each particular case to determine the cheapest type of construc-
tion and most economical spacing of columns. As a general
proposition it has been found that for light loads with ordinary
beam and girder construction the most economical spacing of
columns is 18 ft. each way and for flat slab construction 20 ft.
each way. For heavy loads such as 300 lb. per sq. ft. and over,
it has been our experience that the cheapest colimm spacing for
beam and girder construction is 15 ft. by 15 ft., and for flat slab
construction 17 ft. by 17 ft. In arriving at the most economical
layout it is always well to bear in mind that the construction
which allows the greatest simplicity of form units, together with
the maximum number of repetitions of same, is invariably the
one that will work out cheapest in the end. The fact that the
actual amoimt of concrete or reinforcement required for a certain
floor construction is less than that required in another by no
means implies that this is actually the cheapest floor construc-
tion, as the unit labor of the form work may easily have been
increased out of all proportion.

Digitized by

Google

REINFORCED CONCJRETE CONVENTION HALL AT
BRESLAU, GERMANY.*

By Dr. S. J. TRAUER.f

There is being built at the present time in the City of
Breslau, Germany, a large convention and exhibition hall (Fig. 1),
surmounted by the largest concrete dome in the world. The
building, of reinforced concrete throughout, has a seating capacity
of 9000 persons and standing room for 12,000 people.

The structure (Fig. 2) consists essentially of a main hall,

FIG. 1. — CONVENTION HALL, BRESLAU, GERMANY.

circular in form, connecting directly with four semi-circular halls
called Apsiden, all of which are surrounded by a lower circular
hall serving exhibition purposes. A dome of 213 ft. span with
a rise of 65 ft. rests on the substructure, 65 ft. in height, and
carries a light dome.

The substructure (Fig. 3) consists of four main arches, (A)
of 131 ft. span and 62 ft. rise, circular in plan and supported
by four piers. Each main arch (^4) is supported from the out-
side by four auxiliary arches (B) which cover a smaller semi-

* Translated from author's notes by the Secretary.
fCity Bridge Engineer, Breslau, Germany.

(406)

Digitized by

Google

Traubr on Reinforced Concrete Convention Hall. 407

circular hall called an Apside. These auxiliary arches rest on
individual piers and take the outward thrust from the main
arches. The area of the cross-section of the main arch at the
skew-back is 24 sq. ft. and at the crown 3.6 sq. ft., all reinforced
with round bars of 1.18 in. diameter. The main arches are soUd
and are subject to their own temperature stresses only. The

Smfi^'^^p.

FIG. 2. — PLAN OF CONVENTION HALL, BRESLAU, GERMANY.

horizontal thrust amounts to 771 tons, the pressure at the skew-
back is 1323 tons and in addition the main arches are subjected
to torsion and bending.

The auxiliary arches (JS) which receive about 220 tons thrust
from the main arches, are connected with the main arches and
the small abutments through ball bearings (C) of malleable steel

Digitized by

Google

408 Trauer on Reinforced Concrete Convention Hall.

castings so that the auxiliary arches receive only axial stress.
These arches are 3.28 ft. in width and 5.28 ft. in thickness.

The substructure is entirely independent of the dome itself
and supports the latter by means of steel roller bearings (D) under
each rib, the bearings having radial movement. So that in
general only vertical stresses are transmitted to the substructure,
which is, therefore, not subject to the temperature stresses in the
dome. The wind pressure on the dome is not transmitted in a

FIG. 3. — SECTION THROUGH MAIN AXIS.

radial but in a tangential direction, that is, the direction in which
the four main arches have the highest power of resistance. The
frames (E) over the four abutments of the structures are calcu-
lated to take up the total wind load on the dome.

The dome consists of 32 half ribs (F), which bear on the
top against the pressure ring ((?) and on the bottom against the
tension rmg (H). The pressure ring of 47.2 ft. inside diameter
is surmoimted by an upper light dome (•/)• The pressure ring
with a cross-sectional area of 19.7 sq. ft. carries 551 tons in com-

Digitized by

Goo'gle

Trauer on Reinforced Concrete Convention Hall. 409

FIG. 4. — RIVETED STEEL TENSION RING AND HAUNCH OF ONE RIB.

FIG. 5. — SUBSTRUCTURE AND CENTERING FOR HALF-RIBS OF DOM£.

Digitized by

Google

410 Trauer on Reinforced Concrete Convention Hall.

FI.J. (). — CONCRETING HALF-RIBS OF DOME.

^''«'IIJIil iillllMII'tt'»^"' -

FIG. 7. — VIEW OF INTERIOR.

Digitized by

Google

Trauer on Reinforced Concrete Convention Hall. 411

pression and the tension ring (Fig. 4), with a cross-sectional area
of 29.9 sq. ft., carries 551 tons in tension. The ribs carry a hori-
zontal thrust of 110 tons and 115 tons on the skew-back and
have a cross-sectional area which increases from 3.4 ft. x 2.13
ft. to 3.94 ft. X 2.62 ft. The ribs and the tension and pressure
rings are reinforced with round rods. The ribs are strengthened
by means of rings (M) and horizontal circular slabs (K). The
tension ring is of riveted steel construction.

The roof supports are almost horizontal, the windows (M)
are vertical so that in the case of snow the interior illumination
will not be affected.

The architectural design was made by the City Building
Engineer, Berg; the calculations by the City Bridge Engineer,
Dr. Trauer, and the verification and construction by the firm
of Dyckerhoff & Widmann, of Dresden, Germany.

Digitized by

Google

THE SUITABILITY OF CONCRETE FOR GAS HOLDER

TANKS.

By Herbert W. Alrich.*

In its purpose the gas holder represents the common feature
of all manufacturing enterprises — ^it is the receptacle for the
storage of the product. But in their design, gas holders are not
resembled by any structure or equipment employed in any other
line of industry. A prominent engineering writer has recently
said that the modem gas holder is ''A magnificent achievement
in engineering, and one of the wonders of it is the telescopic
feature." As the majority of engineers are not familiar with
the mechanical features of a gas holder, a description is given;
for in discussing tank design, it is necessary to consider the
structure as a whole.

There are three principle parts to a gas holder; the tank,
the telescopic sections or gas holder proper, and the guide frame.
Each of these parts differs from the others in function and in
the type of construction. The tank, resting upon the ground,
is filled with water up to a level about 15 in. below the top. The
gas holder proper consists of the telescopic sections, of which
there are five in the case of the largest holders that have been
built in this country. These sections consist of cylindrical steel
shells, concentrically located with relation to each other and to
the tank. When the holder contains no gas, these shells are
nested together, resting upon the bottom of the tank and sub-
merged almost completely in the water which the tank contains.
(See Fig. L) The outer-most shell is usually about 3 ft. less
in diameter than the tank, and each succeeding section is about
2 ft. 9 in. less in diameter than the preceding one. All of the
sections are open at each end, with the exception of the inner-
most one, the upper end of which is enclosed by crown plating,
having a spherical form.

The inlet and outlet pipe connections enter the tank through

* Engineers' Department. Coneolidated Gas Company of New York.

(412)

Digitized by

Gqogle

Alrich on Concrete''for Gas Holder Tanks.

413

the bottom and pass vertically upward through the water and
termmate at an elevation above the water and just under the
crown of the inner section. When gas is admitted to the holder,
it first acts upon the crown, lifting that section gradually out of
the water as shown by Fig. 2, until the cup, constructed around
the lower edge of that shell, engages the upper edge of the next

FIG. 1. — CROSS SECTION OF A GAS HOLDER GROUNDED.

outside section. To accomplish this engagement, the upper edge
of each outer section is constructed in the form of a continuous
annular hook, called the grip, and correspondingly the lower
edge of each section, except the outermost one is formed into a
continuous annular cup. Both the grip and the cup are identical
in construction, and differ only in the respect that the grip is
inverted to permit interlocking with the cup as shown by Fig. 3.

Digitized by

Google

414 Alrich on Concrete for Gas Holder Tanks.

As the inflation of the holder continues, each succeeding section
is lifted out of the water, and, in turn, picks up the next outer
section. As each interlocked cup and grip pass upward out of
the tank, there is carried along that quantity of water which is
necessary for forming a hydraulic seal against the maximum

FIG. 2. — HOLDER PROPER JUST BEFORE ENQAQINO AN ADDITIONAL SECTION.

pressure of the holder. This action continues until the holder
is filled to the limit of its capacity as shown by Fig. 4.

While thus ascending, or conversely descending, the sec-
tions are maintained in their relative concentric positions by
two sets of rollers. First, the rollers spaced equi-distant around
the lower edges of the shells, and second, the rollers mounted
on brackets attached to the upper edges of the shells. The first

Digitized by

Google

Alrich on Concrete for Gas Holder Tanks. 415

Digitized by

Google

416 Alrich on Concrete for Gas Holder Tanks.

set are known as the internal rollers, and travel upon vertical
guides, located on the inner surface of the tank and also on the
inner siurface of each shell, except that of the innermost section.
The other set of rollers travel upon the guide rails, which are
carried by the guide frame. The guide frame consists of vertical
columns spaced equi-distant, and about 30 ft. apart around the

FIG. 4. — HOLDER FULLY INFLATED.

circumference of the tank. These columns are connected by
horizontal and diagonal cross bracing. The frame is carried to
such a height as will provide guidance for the inner section in
its maximum upward travel. It will be apparent from this dis-
cussion that the enormous pressure of the wind against the inflated
sections is ultimately transmitted to the guide frame by the tank.

Qigitized by

Google

Alrich on Concrete for Gas Holder Tanks. 417

In the selection of materials in manufacturing, we are sub-
ject to an inexorable law of suitability. In producing any manu-
facture, there will be ultimately employed that material which
possesses preponderant advantage. The use of any other cannot
long be sustained. From the very beginning of the gas industry,
100 years ago, the guide frames and holder sections have been
built exclusively of iron or steel. It does not appear that any
attempt was ever made to use any other materials. In the con-
struction of the tanks, however, many different materials and
combinations of materials have been employed. It is true, how-
ever, that in this country, until recent years, all holder tanks
were built either of brick or stone; but in England, the birthplace
of the gas industry, there were constructed tanks of almost every
conceivable type, many of the designs being more fanciful than
practical. It was inevitable that those constructions without
merit should disappear until the prevailing practice, throughout
the world, during the last twenty years, may be stated as limited
to steel tanks placed upon the ground, and brick or concrete
tanks located below the ground. During the last ten years the
preponderance of advantage in this country has been in favor of
steel tanks for all holders, large or small. In England, though
concrete tanks had been constructed as far back as the year 1870,
the practice of the last decade has been quite imiformly to build
small tanks of steel, and large tanks of brick, there being but few
instances of concrete construction. In Continental Europe,
practice for a decade has followed the English, with the excep-
tion that during the last three or four years there has been a
marked tendency toward the exclusive employment of steel,
though a few concrete tanks of small size have also been con-
structed.

The competitive, economic and geological conditions exist-
ing in England, all favor the brick tank for large holders. While
the population of the Island of Great Britain is somewhat less
than that of the United States, the number of concerns engaged
in holder construction is very much greater, in fact the building
of the largest holders in this country is confined to a very few
companies, having special qualifications and known responsibility.
Hence, the commercial conditions prevailing in this country have
enabled these firms to develop extensive equipments, which have

Digitized by

Google

418 Alrich on Concrete for Gas Holder Tanks.

wonderfully facilitated the construction of steel tanks. When
selecting sites for large holders in England, they appear to have
generally encountered formations of clay possessing remarkable
stability. This has permitted them to construct economical
tanks by lining a cylindrical excavation in the clay with brick
walls, which are thinner than our geological conditions will
generally permit. Unskilled labor is relatively very cheap in
England, and in some cases, the cost of construction has been
partly defrayed by the commercial value of the excavated clay
In the case of the 10,000,000-cu. ft. holder erected at Manchester
in 1909 and 1910, the brick for the tank walls were manufactured
upon the site from the excavated clay. It is a fact, however,
that the time required for the construction of this particular
holder would be regarded as economically impossible in the
United States. The result of all of these conditions has been
that the English holder builders have not developed the special
equipment required for the construction of very large steel
tanks, nor does it appear that the British steel plants produce
plates suitable for the purpose.

To confine the discussion now to American conditions the
controlling requirements in the design of a gas holder are:

1. Structural Stability.

2. Economy.

3. Rapidity of Construction.

4. Durability.

Certain physical phenomena having to do with the distribution
of gas, require the holders to be located upon the lowest avail-
able ground. In this coimtry such sites rarely consist of a very
stable geological formation, and are frequently reclaimed land.
Hence, if under the usual conditions there be undertaken the
construction of an underground masonry tank, there will probably
be encoimtered such diflSculties as to greatly enhance the cost
of the work and prolong the period of construction. If it should
be suggested that a concrete tank might be built above the ground,
a moment's reflection upon its relation to the guide frame is suffi-
cient to dispose of the proposition.

During the last eight years the writer has participated in the
design and construction of the three largest concrete holder tanks
in the world. Two of these tanks are 300 ft. in diameter by 48

Digitized by

Google

Alrich on Concrete for Gas Holder Tanks. 419

ft. 3 in. deep, while the third is 189 ft. in diameter and 41 ft. 6 in.
deep, all inside dimensions. In referring to these particular
tanks, and also throughout the discussion, he wishes to be under-
stood as speaking for himself alone. The work on these three
tanks proceeded in such simultaneous relation as permitted any
improvement in methods developed on one, to be employed to
advantage on another.

At the time the tanks were being designed, Mr. William H.
Bradley, the Chief Engineer of the Consolidated Gas Company
of New York, made a thorough inquiry as to the design of any
other concrete tanks that had been constructed anywhere in
the world. There was foimd little precedence to consider. As
regards the two 300-ft. tanks, the undertaking was one of unusual
n^agnitude, for the holders were the largest that have ever been
built. The possibilities of nickel steel were not at that time
fully understood, hence the construction of steel tanks would
have been unthinkable, as there would have been required plates
4 in. thick, connected by rivets 3 in. in diameter. It was evident
that masonry construction of some kind must be employed. A
number of different designs were worked out, analyzed and com-
pared, resulting in the conclusion that the most advantageous
construction would be a plain annular wall of reinforced concrete,
and all three tanks were thus built. The work upon all three
tanks was accomplished with entire success. While there were
special conditions, the controlling factors in these instances,
and determining the selection of concrete construction, the
writer concludes that steel is far preferable, as a general proposi-
tion. The smallest of these three tanks required a full year for
its construction. An entire holder of the same size having a
steel tank, can be constructed in seven months, without special
effort. It has also been demonstrated to the satisfaction of the
writer, that in no case where a steel tank is at all possible, will
the cost reach 70 per cent of the outlay required for a correspond-
ing concrete construction.

The question as to whether steel or concrete tanks are the
more durable, must be settled in another generation; but the
writer believes that the steel tanks are the more expedient. He
has knowledge of several brick tanks having been badly damaged
by blasting in their vicinity. From his knowledge of other con-

Digitized by

Google

420 Alrich on Concrete for Gas Holder Tanks.

Crete structures having been badly cracked in a similar way,
he must conclude that concrete tanks are not immune from the
hazard. The durability of concrete tanks is also open to question,
in view of the peculiar behavior of concrete observed by the
writer, in the instance of two of the most prominent engineering
undertakings in the United States, in which cases, several years
after the concrete had set, masses of it became soft and pulpy,
requiring replacement. A holder tank is not such a structure
as one upon which such repairs can be easily made, but it is one
of such importance as to forbid speculative constructions.

The writer has examined and reported upon a large number
of steel tanks, some of which had been in service for over twenty
years. While these older tanks are doubtless constructed of a
steel far inferior to that which is now produced, all of these tanks,
with one exception, could survive the youngest man now engaged
in the gas business. It has sometimes been urged against the
durability of steel tanks, that all favorable conclusions had been
formed from exterior examinations. The writer has had the
opportunity of examining internally, some of the oldest steel
tanks in existence, through which he concludes that, excepting
at the water line, there is no perceptible deterioration. In one
case, the holder had been employed for twelve years, in connec-
tion with a process of gas manufacture now obsolete. As a part
of this process, a non-luminous gas containing some sulphur
was stored in this holder, which resulted in the sulphur being
absorbed by the water in the tank. The writer found the interior
of this tank to be but very slightly pitted.

The writer would state it as his conviction, that while not
possessing one single element of advantage over steel tanks,
that concrete tanks are subject to the following comparative
disadvantages:

1. Increased cost, generally 75 per cent,

2. Longer period required for construction, generally 100
per cent for the entire holder.

3. Liability to impairment from unforeseen or unavoidable
causes, such as internal stresses, bad water, oil or alkaline soil
and blasting.

4. Difficulty of making repairs and obtaining the original
strength in case of rupture.

Digitized by

Google

AuacH ON Concrete for Gas Holder Tanks. 421

5. Possibility of corrosion of reinforcement, the vital ele-
ment of the tank's strength.

6. Inaccessibility of metal to inspection.

7. Greater load on fomidations.

8. Requires the most rigid inspection and supervision to
control the quality of the work during construction.

9. Liability to serious dfficulties from storms during con-
struction, due to the depth of excavation, the requirement of
maintaining adjacent streets, requiring sheet piling 50 ft. deep
and massive shoring timbers. Also liability of damage to the
concrete work itself from the same cause.

In submitting a discussion of this question to the American
Gas Institute in 1910, the writer urged the following as the twelve
distinct advantages of steel tanks when compared with those of
masonry:

1. Less cost.

2. Shorter period required for construction.

3. The ease with which the quality of the work may be con-
trolled during construction.

4. The high state of development in fabrication and erec-
tion.

5. Susceptibility to exact computation and greater reliability
under stress.

6. Accessibility for inspection.

7. Tank may be placed at any elevation with relation to
the ground line that may be desired.

8. No liability to damage by storm durmg construction.

9. No internal stresses from shrinkage or temperature that
are serious.

10. Possibility of rectifying an imequal settlement.

11. No liability to cracking from undetermined causes.

12. The ease of making repairs and obtaining the original
strength.

Every one of the enumerated disadvantages of concrete
tanks, and advantages of steel tanks might be separately ex-
tended and elaborated, but the writer will confine himself to
items 2 and 4 under steel tanks.

The bottom of a steel tank, excepting an outer course, con-
sists of rectangular steel plates, generally about f in. thick, con-

Digitized by

Google

422 Alrich on Concrete for Gas Holder Tanks.

nected by single riveted lap joints. The outer course consists
of heavy segmental plates, which conform the bottom to the
circle. Riveted around the outer edge of this outer course, is
the bottom curb, which, in the case of a large tank, will be an
8 X 8 X IJ in. angle. The bottom course of curved plates, form-
ing the shell of the tank, is riveted to the upstanding leg of this
curb angle. The next course of plating is then attached to the
lowest course by a single riveted lap joint running horizontally
around the tank circumference. The other courses are connected
in a similar manner, the entire number in the height of the tank
usually being about eight or nine. It is necessary for the vertical
joints, which occur about 30 ft. apart around the circumference,
to be spliced in a manner capable of resisting the full circum-
ferential tension. To accomplish this, the joints are covered
by double butt straps, each such splice being quadruple riveted
with double shear rivets, and triple riveted with single shear
rivets. The total number of rivets in the cylindrical shell of the
tank alone, may be as many as 25,000.

After the bottom of the tank has been completed, a vertical
steel post is erected at its geometric center. A circular rail is
also laid concentrically upon the tank bottom. Mounted upon
trunnions, upon the top of the central steel post, are two radial
traveler arms, the outer ends of which are supported by legs
having wheels running upon the circular rail. These radial
travelers are equipped with such suitable attachments, as will
permit one of them to be used for assembling the curved plates
while the other supports the hydraulic-pneumatic riveting
machine. Thus, as one traveler proceeds around the circumfer-
ence lifting the plates into position, it is followed by the other
with the riveter. Quite obviously, the first traveler is able to
keep well ahead of the riveting, which permits that traveler to
participate in the simultaneous erection of the five-holder shells.
When the tank and the holder sections have been completed,
the tall center post around which the travelers revolve, is replaced
by a much shorter one, resting upon the center of the crown,
while the outer leg of one of the travelers has its lower section
removed, and thus shortened, it travels upon a circular rail which
has been attached to the completed crown. The other radial
traveler is entirely removed. After water has been placed in the

Digitized by

Google

AuacH ON Concrete for Gas Holder Tanks. 423

tank, the holder is gradually inflated with air, and as it rises the
remaining radial traveler moves aromid the circumference, erect-
ing the guide frame successively in tiers. It requires but little
reflection to perceive that if the holder is to be provided with a
concrete tank, any such co-ordination is impossible. As to the
time required, it may be stated that in the instance of the largest
steel tanks ever built, which were 251 ft. 3 in. diameter, the lower
courses of which consisted of 2]^-in. plates and having 26 double
butt joints in each course, the tanks were erected and riveted
complete at the rate of 32 hours per course. One of these holders
was entirely completed in 7j months. If there be a method of
concrete tank construction which would have permitted the
completion of that same holder in two years, it has not yet been
found.

In closing, the writer will condense into a few words, his
opinion on the matter as stated in another discussion. "He may
be charged with entertaining a strong prejudice against masonry
tanks. He will admit that he is opposed to any type of tank,
25 to 40 per cent of the cost of which may go into digging a hole
in the ground instead of putting quality into the structure.''
To this he would add his conviction that a site which forbids a
steel tank is not the place to put a holder.

Digitized by

Google

PROTECTION OF STEEL IN CATSKILL AQUEDUCT
PIPE SIPHONS.

By Alfred D. Flinn.*

Cement manufacturers and many cement users are already
so familiar with the Catskill aqueduct which New York City is
building to convey an additional supply of water from the Catskill
mountains that a general description is not here necessary. This
aqueduct is to have a nominal capacity of 500,000;000 gallons
daily. In the 92 miles of its length, it crosses 14 minor valleys
where metal pipe siphons were determined upon in preference to
reinforced concrete pipes or deep pressure timnels in rock; three
of these siphons are west of the Hudson river, and eleven east;
in seven of them the diameter of the steel shell is 9 ft. 6 in.; in
four this diameter is 9 ft. 9 in., and in the remaining three, 11 ft.
3 in. These various diameters were determined by an economic
distribution of the available fall or head. The length of the
siphons varies from 608 to 6671 ft.; one of the U-ft. siphons
is 5584 ft. long; the total length of all siphons is 33,031 ft. The
thicknesses of plates are i^-, i-, xk-i H" a^^d i"^^- Th^ mairimum
heads on the siphons range from 50 to 340 ft. With one excep-
tion, each siphon rises to the hydraulic gradient at each end and
is there connected by means of a concrete chamber to the adjacent
portions of the cut-and-cbver or tunnel aqueduct. They are all
of open hearth steel. Ultimately there will be 3 pipes in each
siphon, in order that the siphons may have the full capacity of
the aqueduct and also provide for one pipe being temporarily
out of service for cleaning, repairs or renewal. Only the middle
pipe of each siphon is being laid at this time. The 7 more northerly
siphons are included in Contract 62 and the 7 southerly siphons
in Contract 68.

For years, both in connection with the work of the Board
of Water Supply and in other engagements, several of the Board's
engineers have been observing the results obtained by coating
steel pipes with the asphalt, tar and other dips commonly em-

* Department Engineer, Board of Water Supply of the City of New York.

(424)

Digitized by

Google

Flinn on Protection op Steel Pipe with Cement. 425

ployed. Study had also been made of some special forms of
protection of steel pipes. Excepting Portland cement grout
or mortar used in one or two cases, in a small way, none of the
coatings of which knowledge could be had gave evidence of real
permanence, nor have they been fully satisfactory in other respects;
therefore it was decided to jacket the steel pipes of the aqueduct
outside with rich concrete, and to line them with Portland cement
mortar. The outside concrete was to have a minimum thickness
of 6 in., and the mortar lining a thickness of not less than 2 in.;
after some experimentation, these dimensions were definitely
adopted, as shown on Fig. 1, which is a reproduction of the draw-
ing of the standard types of pipe construction.

Means for applying the concrete and mortar to the pipes
so as to attain intimate, complete and permanent adhesion, in
spite of the unavoidable distortions due to handling, tempera-
ture changes, and the water stresses, were studied, along with
the most economical and feasible methods for the various steps
in construction. Preparation of the steel was also carefully
studied. Numerous experiments on a small scale, and finally
on substantially full-size, were conducted. For the latter, a steel
pipe 9 fib. in diameter and 12 ft. long, of |-in. riveted plates, was
lined by plastering and by pouring grout or very thin mortar
into the space between a cylindrical form and the inner surface
of the pipe. In the plastering experiments several kinds of metal
reinforcement were tried; also, terra-cotta and cement blocks
or tiles were bedded on mortar and plastered. Briefly, it may
be stated that no combination of plasterer's skill, with the various
materials suggested, gave linings that were adequate and this
method was expensive. The methods using sohd*tiles or blocks
were more successful than those using any form of metal lath.
But when removed, all the plaster coatings showed a tendency
to separate at the surfaces between the successive layers. Grout-
ing proved by far the most satisfactory and least expensive,
and was adopted as the basis of the contracts.

In fabricating the pipes, the plates were bevel-planed on
their edges, then punched for the rivets and then each plate was
bent to proper radius by bending rolls. This bending cracked
the mill scale and removed a considerable portion of it. Pickling
was resorted to for the removal of the remaining mill scale, the

Digitized by

Google

426 Flinn on Protection of Steel Pipe with Cement.

K
o
Z

Digitized by

Google

Flinn on Protection op Steel Pipe with Cement. 427

rust and the dirt. Two vertical wooden tanks of suitable diam-
eter were placed side by side, one containing hot dilute sulphuric
acid kept at a strength of about 5 per cent of oil of vitriol, which
was approximately 93 per cent pure sulphuric acid; after 15
minutes in this solution at a temperature of about 125 deg. F., each
plate was of a uniform clear steel gray color all over and when
removed was at once dipped into clean hot water in the adjacent
tiank. Riveting followed pickling quickly, the pipes being made
in 15-ft. sections of 2 rings of 7^-ft. net length each. Each ring
Was one plate, except in the 11-ft. 3-in. pipes, and in those made
of plates thicker than ^ in., for which two plates were used. It
has been observed that where steel is exposed to corrosion with
the mill scale on the surface, the corrosion tends to concentration
at certain points and is accompanied with pitting; hence the
care exercised to remove the mill scale thoroughly.

Observation and such information as was available indicated
that cement mortar adhered most strongly to steel and afforded
the best protection when it could be applied directly to the clean
surface of the metal. Having obtained a satisfactorily clean
surface in the shop, there remained the problem of preserving this
surface in as good condition as practicable until the pipes could
be laid, covered with the concrete, and lined with the mortar.
Various temporary coatings were suggested, but it was finally
decided to use whitewash, and so each pipe was given a coat of
heavy lime whitewash before it left the shop. To each barrel
of whitewash, about 50 gal., there were added about 20 lb. of glue;
the glue was dissolved in water before mixing in the whitewash.
After this mixture had been used for a time, about 1 lb. of Port-
land cement was added for each gallon of whitewash. This
whitewash, applied with brushes, did not adhere very well, and
through lack of care in handling, the pipes suffered more or less
almost as soon as it was applied. They suffered more from expo-
sure to the weather, however, than from abrasion; but even
where the whitewash was not disturbed light rusting occurred.
Only around the rivets and at joints where the whitewash had
formed a very thick coating was there no sign of rusting. Hence,
as delivered at the trench, the pipes had more or less complete
coats of light yellow rust very uniformly distributed on the bared
portions of the steel, without indications of any tendency to

Digitized by

Google

428 Flinn on Protection op Steel Pipe with Cement.

pitting. This light rust has been regarded as not seriously
objectionable.

Probably the greatest obstacle to securing the desired pro-
tection of the steel between the time it leaves the shop and the
time it is covered with the mortar or concrete is the tendency of
the steel mill and fabricating shop to push their operations mu^h
more rapidly than the pipe can be laid, tested, and covered^ and
the uncertainties of the field work, which make it desirable to
have pipe on hand ahead of the progress of laying. These cir-
cumstances resulted in some cases in the gradual formation of
heavier rust; this has been removed, as required by the specifica-
tions, as is also the whitewash which still adheres to the pipes,
just before the applications of the jacket or lining.

Ease of removal and probable lack of objectionable effects
upon the mortar, if small quantities should not be removed,
were among the reasons for adopting whitewash. Unquestion-
ably, Portland cement grout would have stuck more tenaciously,
but it would have been correspondingly difficult to remove, and
it was thought probable that the mortar of the lining and jacket
would not adhere well to the old cement surface; these were
considered sufficient arguments for forbidding its use. For
removing the rust and dirt from the pipes, wire brushes are com-
monly used, and in some of the worst places steel scrapers also.
Inside some of the siphons the surfaces have been rubbed with
empty cement bags after the wire brushing. This final cleaning
is done in short stretches just in advance of the placing of the
concrete or mortar.

To support the pipe in the trench so as to permit the placing
of the concrete jacket beneath it, and also aid in bringing the
pipe to line and grade as it was being laid, concrete blocks called
cradles were built in the bottom of the trench. Figs. 2, 3 and 4
show the several styles. At first attempts were made to have
these cradles fit the bottom of the pipe closely, but this gave
trouble and the shape was modified. Some cradles were made
about 24 in. square and extended about 6 in. below the ordinary
sub-grade of the trench, with their tops at the proper grade for
the bottom of the pipe, and, on the whole, this shape was satis-
factory. Some of the longer cradles were cracked, due probably
to unequal bearing or shocks in placing the pipe. Most of these

Digitized by

Google

Flinn on Protection op Steel Pipe with Cement. 429

cracks were due primarily to the uneven bearing of the pipe on
the cradles.

The pipe having been laid, riveted and calked was filled
with water to hydraulic gradient, inspected, and leaks further
calked. Bulkheads were placed in the open ends of the pipe and
a small riser pipe carried up to the proper elevation to represent

IS'

-W'

FIG. 2. — CONCRETE CRADLES FOR PIPE SIPHONS.

the working head when the aqueduct would be in service. To
maintain this pressure on the pipe constantly, a small reservoir
or tank was attached to the top of the riser pipe and kept just
spilling over. While the pipe was still full of water under this
normal working pressure, the concrete jacket was placed about
it by methods similar to those used in building concrete conduit,
excepting, of course, that no inside form was needed, Figs. 5 and

Digitized by

Google

430 Flinn on Protection of Steel Pipe with Cement.

6. The water pressure was continued until this concrete had
attained considerable strength, the period depending upon the
weather, kind of cement, and other conditions. It was found
undesirable to maintain the water pressure by direct pumping
into the pipe while the concrete was being placed and was harden-
ing, since fluctuations of head caused a few cracks. Maintaining
this pressure by the small overflow tank and riser pipe mentioned
above was more satisfactory. When the last concrete was suffi-

(//? roc/r on/yj /or /79/€fyo//ff

FIG. 3. — TYPICAL ARRANGEMENT OF CONCRETE CRADLES AND JOINT HOLES FOR

FIELD RIVETING.

ciently hardened, the water was slowly withdrawn from the
pipe.

Three variations in the procedure of placing* the concrete
jacket were tried: (l) Monolithic, as specified; (2) first the invert,
then the remainder, and (3) first invert and side walls to the hori-
zontal diameter, and then the arch. The best contact between
the jacket and the pipe seemed to have been obtained when the
concrete was placed monolithically. It was found unwise to
permit the concrete to be dumped from the buckets with a greater

Digitized by

Google

Flinn on Protection of Steel Pipe with Cement. 431

drop than 2 ft., because the vibration tended to crack the concrete
recently placed.

On one siphon the average rate of placing concrete jacket
was about 130 ft. a week; on another, 170 ft.; and on a third,

PIG. 4. — LAYING STEEL PIPE ON CONCRETE CRADLES.

190 ft., the maxima on these three siphons being, respectively,
223 ft. per week, 297 ft. and 260 ft. On one siphon the progress
of 30 ft. of full section per 8-hour shift, using about 50 yd. of
concrete mixed by machine was attained. Many factors influ-

Digitized by

Google

432 Flinn on Protection of Steel Pipe with Cement.

enced progress, such as the method of mixing, method of trans-
portation whether by cableway, wheelbarrows or cars, and the
style of outside forms, whether steel or wood, and the conve-
niences for moving and setting the forms. Earth covering was
placed in many cases immediately after the completion of the
concrete jacket, but, in a few instances, considerable time elapsed
. before the concrete was covered. So far as observed, there have
been few cracks in the concrete jackets built to the end of last
season. •

FIG. 5. — METHOD OF PLACING CONCRETE JACKET AROUND STEEL PIPES.

Two quite different methods were at first tried by the two
contractors for lining pipes. Under Contract 62, the contractor
began at once by grouting with forms; under Contract 68, an
attempt was made to use the cement gun, and the Hunter's Brook
siphon, 1493 ft. long, 9 ft. 9 in. diameter of pipe shell, was lined
in this way. The cement gun* was fully described in Engineer-
ing Recordf July 1, 1911, and other descriptions have been pub-
lished, so that it need not be described again in this paper. Several

* See Proceeding*, Vol. VII. p. 504.— Eo.

Digitized by

Google

Flinn on Protection op Steel Pipe with Cement. 433

machines are shown m Fig. 7. Charges of dry sand and cement
mixed in prescribed proportions were placed in the charging
chamber of the machine and then dropped into the pressure
chamber. From the latter the mixture was rapidly discharged
under an air pressure of 50 to 60 lb. per sq. in. through a rubber
hose, while through a parallel hose joining the former in a special
nozzle, water under pressure was discharged so that the sprays
of water and of sand and cement were commingled. A pressure
of about 30 lb. was maintained at the nozzle and so the mixture

FIG. 6. — ONE OF THE CONCRETE MIXERS.

was thrown with considerable force against the surface being
coated. This aided in securing an excellent union between the
mortar and the steel and more especially between the successive
portions or layers of the lining. The operator held the nozzle
2 or 3 ft. away from the surface being coated, moving it back
and forth continuously, meanwhile controlling the discharge from
the nozzle by lever valves.

The force with which the mixture of sand, cement and water
is thrown against the surface being covered causes a measurable

Digitized by

Google

434 Flinn on Protection of Steel Pipe with Cement.

proportion of the sand grains to rebound from the surface and
fall into the bottom of the pipe, particularly when begiiming the
first layer on the bare steel. By analysis this dry material was
found to contain about 1 part of cement to 3f parts sand; it was
collected and used for making the invert or bottom part of the
pipe lining, which was deposited as a mortar and screeded to
shape in advance of the gun work. Because of this separation
of sand, an excess of it is put into the dry mixture in order to
prevent the lining being richer than intended. Another result

FIG. 7. — FOUR CEMENT GUNS ON HUNTER's BROOK SIPHON.

is that the atmosphere inside the pipe is commonly very dusty
and in order to make reasonable working conditions for the men
artificial ventilation was resorted to. Furthermore, the mortar
thus applied contained less water than the grout, and more atten-
tion must be given to keeping the lining moist in order to minimize
shrinkage cracks. Fine cracks have formed rather numerously
and are probably due, in considerable measure, to the failure
to keep the lining wet enough while it was setting and hardening.
In spite of the difficulties, a lining which up to date seems satis-
factory was secured. When this siphon was completed, the use

Digitized by

Google

Flinn on Protection of Steel Pipe with Cement. 435

of the cement gun was discontinued, the contractor giving as a
reason excessive cost. Lining by the grouting method was then
adopted under Contract 68.

It has been common practice to place the lining in the invert
of the pipe by screeding, the width thus placed varjring from
about 2 to 6 ft. measured along the arc. Practical difficulties
prevented the use of a simple complete cylindrical form; among
th^e difficulties were going around curves, either vertical or
horizontal, collapsing one set of forms sufficiently to pass it for-
ward through another set which must remain in place while the
grout was hardening, and cleaning, lubricating and inspecting the
outside of the form inside the pipe. Consequently, forms in
panels about 2 ft. wide and 15 ft. long (7^ ft. at curves) were
adopted. These are supported on wooden ribs or centers adjusted
to give the proper thickness of lining and firmly braced, Fig. 8.
The lining has been commonly poured in 15-ft. sections in one
operation.

The grout is poured from outside the pipe through a 2i-in.
wrought-iron pipe secured into a rivet-passing hole. In the
Southern Department this pouring pipe is at the downhill end
of the section and is long enough to give a head of about 4 ft.
on top of the uphill end. A vent pipe is fastened in the uphill
rivet-passing hole of the section, the bulkhead forming the end of
the lining being placed just below it. For poiu'ing grout the con-
tractor for the northern siphons has always used two mortar
boxes set on a temporary staging over the upper end of the sec-
tion to be grouted. The mortar is mixed to the proper consistency
in alternate boxes and is allowed to flow into the pipe through a
hole controlled by a sUding wooden gate. All mixing is done by
hand, and materials are carefully graded. It generally takes
about two hours to fill a section, after which a man is kept on for
an hour or two in order to feed in sufficient grout to get the desired
consistency. There is a noticeable tendency to get a porous or
thin condition at the upper end of the section, near the grouting
hole. To avoid this, the riser is removed two or three times dur-
ing the pouring and the thin material which collects at the top is
allowed to escape. The riser is then put back and grout added
until desired results are obtained. In some cases it takes nearly
two hours after the main operation to get grout of proper con-

Digitized by

Google

436 Flinn on Protection of Steel Pipe with Cebient.

sistency at the foot of the riser. At the finish of the pouring a
small pipe is inserted through the large pipe, to permit the escape
of the last air while grout is poured through the larger pipe and
churned into the small remaining space, to insure complete fiUing.
For the first batches, or nearly up to the horizontal diameter of

FIG. 8. — WOODEN FORMS IN PANELS FOR PLACING CONCRETE LINING BY THE
GROUTING METHOD.

the pipe, the grout is mixed 1 part cement to 1 part sand, and the
remainder about 1 to 2.

Some fine cracking has occurred in the lining placed by the
grouting method as well as in that deposited by the cement gun.
None of this cracking is believed to be serious. As an aid to
preventing cracks and a safeguard against the remote possibility
of small pieces of the lining becoming loose and falling out, both
contracts provide for using wire fabric in the mortar, as a rein-

Digitized by

Google

FiiiNN ON Protection of Steel Pipe with Ceiosnt. 437

foroement, but up to the end of last year none had been used.
Permanent absolute adhesion of the concrete or mortar to the
steel IS not being attained at all parts; this has been proven by
careful sounding with a hammer, but on cutting into hollow sound-
ing spaces, the space between the mortar or concrete and the
metal has been found almost infinitesimal in width. Before
preparing the contracts, such an occurrence was anticipated and
some tests were made at the Board's laboratory to help in deter-
mining how serious a matter this would be. A brief statement
of these experiments follows:

One experiment was made as follows: Four circular slabs
15J in. diameter and 3 in. thick were made of concrete in pro-
portions 1 cement: 2.7 Jerome Park screenings: 6.3 Jerome Park
stone (gneiss) by weight, the consistency of the mix being rather
dry. Carefully imbedded in the center of the slabs were four
i-in. round, softnsteel rods spaced 3 in.* apart, so that there was
at least IJ in. of concrete in all directions from each rod. When
the concrete had set, two of these slabs were immersed in water
in two tanks 1 ft. 6 in. in diameter and 2 ft. deep. The other two
slabs had two galvanized-iron cylinders 15^ in. in diameter
cemented to them, and a head of 20 in. of water maintained on
them. The water in the tanks and cylinders was kept at a constant
depth and head by the addition of water whenever it was neces-
sary. The slabs subjected to percolation leaked rapidly at first,
but became gradually tighter, and during the last few months
of the tests there was very little leakage. The tests were con-
ducted in open air, and were subject to the variations in outdoor
temperature. The tests commenced July 13, 1907. On March
23, 1909, one year and eight months later, the slabs were broken
and rods examined for corrosion. It was found that the pro-
tection of the rods by the concrete was perfect in all cases, there
being no sign of corrosion on either the black finish of the metal
left by the rolls or on the bright ends of the rods exposed in cutting
to length with the hacksaw. The concrete when broken was
found to be thoroughly saturated, showing that the water had
full access to the rods.

Another experiment was conducted as follows: Six steel
plates 8 in. by 16 in., of 12 gauge, cleaned by pickling and then by
rubbing with emery cloth, were placed horizontally in a galvanized

Digitized by

Google

438 Flinn on Protection op Steel Pipe with Cement.

iron tank 18 in. in diameter and 24 in. deep, separated from the
bottom by two IJ-in. bars of alberene stone, and from each other
by i-in. wood dowels. The first pair of plates was put in with-
out any protective covering. The second pair had their upper
surfaces protected by a slab of cement mortar 2i in. thick, not in
contact with the steel, but separated from it by two metal strips
about .04 in. thick. The third pair of plates was protected by
cement mortar slabs 2 in. thick, cast directly on the steel, and
apparently adhering to it firmly. The tank was then filled with
proton water to a depth of 4 in. above the top of the uppermost
mortar slab and kept filled the entire duration of the test, the
water being renewed twice monthly.

After two years' immersion, the plates were taken out and
cleaned oflf by washing with a sponge. The first pair of plates
showed heavy corrosion. In numerous places the entire layer of
oxide had separated from the steel and formed blisters, leaving
the bright steel surface underneath. The second pair of plates
showed a very slight corrosion. Most of this washed off, thus
indicating a considerable protective infiuence of the mortar slabs
even when separated from the metal by a space of .04 in., open
at the edges all around. When the mortar slabs were removed
from the third pair of plates, it was found that a part of the sur-
face of the steel had a distinctly different appearance from the
other part. One part was clean and wet; the other part was
covered by strongly adhering particles of mortar and was dry,
thus indicating that there had been actual adhesion of the mortar
only over the latter part of the surface and that the former had
been separated by a space large enough for the water to enter.
There had, however, been no rusting except at some places near
the edges of the wet part of the surface, where apparently the
space had been big enough to allow circulation of the water.
The water in the rest of the space had evidently been so highly
charged with lime that no corrosion could take place.

Digitized by

Google

Flinn on Protection of Steel Pipe with Cement. 439

Extracts from Contract 62.
general sections.
Order of Work,
Sec. 19. The steel pipe shall be laid, tested, and made tight against
hydrostatic pressure; then surrounded by concrete while still under the normal
hydrostatic pressiu-e, after which the mortar lining shall be placed. Except
at overhead stream crossings, no stretch of pipe shaU be left not well protected
from frost between stretches of pipe around which concrete has been placed.
As soon as practicable after the concrete covering and mortar lining have been
placed, the pipe shall be covered with earth; and all pipe covered with con-
crete, whether or not mortar lining had been placed in it, shall be protected
in freezing weather by at least one foot of earth.

MORTAR UNINO FOR STEEL PIPES.

(Item 26.)
Work Included.
Sec. 26.1. Under Item 26 the contractor shall build a mortar lining
inside the steel pipe and cast-iron bell castings as specified, directed or
approved.

Description and Proportions,
Sec. 26.2. The lining shall consist of Portland cement and sand, mixeJ
in ordered proportions, probably one part of cement to two parts of sand
The quality of the sand shall be as specified under Items 28 to 30. Reinforce-
ment, if used, will be paid for under Item 27. The lining shall be of sub-
stantially uniform thickness throughout the entire circumference except for
the unavoidable variations due to lap of the plates, butt straps, and rivet
heads. The thickness over the inner course of plates is to be 2 in., that is,
the internal diameter of the lining shall be 4 in. less than the nominal diameter
of the steel shell.

Forms.
Sec. 26.3. Forms shall be of steel, or of wood covered with galvanized
sheet steel, and shall be especially constructed so as to have sufficient strength
and yet be adjustable so as to give a uniform space between them and the shell
of the pipe. Great care shall be exercised to secure forms which will leave the
surface of the lining perfectly smooth. Forms which give unsatisfactory
results after use shall be satisfactorily repaired or replaced. The length of
the sections of forms will not be restricted provided satisfactory means are
adopted for controlling the uniform thickness of the lining and the correct
spacing of the reinforcement, but sufficient sections shall be provided adapted
for lining the pipe on curves, where lining shaU be placed in sections about 7 ft.
long. Each time the forms are used they shaU be thoroughly cleaned and then
coated with some approved inadhesive substance which wiU prevent the mortar
from sticking to the forms without injuring the mortar. The lining of manhole

Digitized by

Google

440 Flinn on Protection op Steel Pipe with Cement.

castings, and of the blow-off elbows to the sockets, shall be monolithic with
the lining of the steel shell, and forms shall be so constructed as to permit this
method.

Method of Placing.
Sec. 26.4. The lining shall in general be placed by pouring a grout
around an internal form, through holes cut in the top of the pipe for that pur-
pose in the manner and at locations specified in Section 19.11. Sections shall
be so terminated as to bring a hole at the upper end which shall be arranged as
an air vent. The mortar shall be mixed to a thick creamy consistency and
allowed to flow into place as uniformly as possible. When the section is filled
to the top, the pouring of the grout shall be continued until grout runs from
the air vent; then headers of steel pipe shall be screwed into the inlet and out-
let holes and filled with grout so as to put a head of at least 4 ft. on the highest
part of the section, and these headers shall be kept filled with grout until
the grout has set, when the pipe shall be removed and the hole made water-
tight by a screw plug. During the pouring of the grout, the form shall be
tapped to loosen air bubbles, and a careful watch shall be maintained to pre-
vent leaks. The work shall be so planned that the grout can be poured con-
tinuously from start to finish of the section. Any interruption greater than
fifteen minutes, whether due to leaks or any other cause, may be sufficient
reason for the rejection of the entire section.

Lining with Plaster.
Sec. 26.5. Should there be any portion of the interior of the pipe
which it is impracticable to line by grouting, this portion shall be lined by
plastering the pipe with mortar, mixed as specified in Section 26.2. Only
skilled masons or plasterers shall be allowed to do this plastering, and a section
once started shall be prosecuted imtil finished, with only such pauses as are
necessary for a sufficient setting of a layer to permit the next layer to be placed.
Each layer shall be as thick as is feasible to apply, so that as few layers as
possible may be necessary. The surface of each layer, except the final one,
shall be brushed to thoroughly remove the laitance and then deeply scratched
or otherwise satisfactorily treated to give a bond with the succeeding layer.

Removal of Forma.

Sec. 26.6. The forms shall be removed within twelve houTB of the time
set by the engineer, and the section, if accepted, shall receive immediately
such repairs as required, in the manner directed. It is possible that the lower
part of the Lining will, in many cases, show a sandy surface, and if so, it shall
be brushed with enough neat cement wash to fiU the pores and no more, and
troweled to a smooth finish. Any section not accepted shall be immediately
removed by the contractor at his own expense and replaced by acceptable
lining.

Prevention of Freezing; Bvlkheada.

Sec. 26.7. Suitable bulkheads shall be erected in the pipe to prevent
freezing inside the pipe cither during the placing of the Lining or after its

Digitized by

Google

Flinn on Protection of Steel Pipe with Cement. 441

completion. They shall be removed before the completion of the contract,
if ordered. Lining shall not be placed in the uncovered pipes over streams
during, or within a month before, freezing weather, unless the pipe is satis-
factorily protected, and, after lining these portions of the pipe, water shall not
be allowed to stand and freeze there.

Measurement and Payment.

Sec. 26.8. For placing the lining in the steel pipe and cast-iron bell
castings, the contractor shall receive the price per linear foot of pipe stipulated,
the measurement to be made along the axis of the pipe, this price to include
all labor and materials necessary to complete the lining in a thorough and
approved manner except only that the cement required will be paid for under
Item 35 (Portland cement).

REINFORCEMENT OF MORTAR LINING FOR STEEL PIPES.

(Item 27.)

Description.

Sec. 27.1. Reinforcement may be ordered under Item 27 for any part
or the whole of the mortar lining. The reinforcing material shall be galvanized
steel mesh of a style and weight approved, provided, however, that no rein-
forcement shall be required of which the lowest price obtainable by the con-
tractor, f .o.b. New York City, exceeds f cent per square foot, for lots of 10,000
sq. ft.

Placing.

Sec. 27.2. The reinforcement shall be placed approximately in the center
of the mortar lining. The reinforcement may be kept away from the pipe
by distorting the reinforcement at frequent intervals, so as to make points
projecting toward the pipe. Unless otherwise permitted, small blocks of
mortar shall be attached to the reinforcement, for the purpose of keeping it
away from the form. Metal shall be lapped at least 6 in. at all longitudinal
joints.

Measurement and Payment.

Sec. 27.3. The quantity to be paid for under Item 27 shall be the
number of square feet of lining, meastu^ as of a mean diameter 9 ft. 4 in.,
in which reinforcement has been ordered and placed. This does not include
any allowance for lap. The price stipulated shall include the cost of the
reinforcing metal, royalty if any, cutting, shaping, bending, wires, clips, mor-
tar, and other devices used for holding the reinforcement in place, or for splic-
ing the strips; and it shall further include any additional expense of forms,
tools, appliances, and labor other than the expense that would be required
for finishing the mortar lining under Item 26 without reinforcement. .

Digitized by

Google

442 Flinn on Protection of Steel Pipe with Cement.

The gain in smoothness of interior by covering the rivet
heads and the plate laps has been computed to so increase the
hydraulic capacity that three pipes equal four without lining.
The total cost for the siphons with three lined and jacketed
pipes is estimated at about the same as for four pipes constructed
and coated in the more usual way. Obvious incidental advantages
are secured by the more permanent construction.

The Board of Water Supply of the City of New York is

Table I. — Contract Prices for Steel Pipes, Concrete Jacket and

Mortar Lining.

Description.

Contract 62.

9 ft. 6 in. steel pipe, ^/u-in. plate, lap jointed

ft. 6 in. steel pipe, V't-in. plate, lap jointed

9 ft. 6 in. steel pipe, Vi-in. plat«, longitudinal
jointed

9 ft. 6 in. steel pipe, */ia-in. plate, longitudinal i
jointed

9 ft. 6 in. steel pipe. "Ae-in. plate, longitudinal

jointed

9 ft. 6 in. steel pipe. */4-in. plate, longitudinal seams butt-
jointed

Mortar lining for steel pipes

Reinforcement of mortar lining for steel pipes

Concrete masonry around steel pipes

Portland cement

butt-
biitt-
biitt-

Contract
Price.

Contract Sfi.

9 ft. 9 in. steel pipe, "'kt-ia. plate, lap jointed

II ft. 3 in. steel pipe, ^/it-in. plate, lap jointed.. . .
11 ft. 3 in. steel pipe, Vrin. plate, lap jointed. . . .
11 ft. 3 in. steel pipe, ^/a-in. plate, longitudinal

butt-jointed |

11 ft. 3 in. steel pipe, */i6 in. plate, longitudinal seams

butt-jointed '

Mortar lining for 9 ft. 9 in. steel pii)e '

Mortar lining for 11 ft. 3 in. steel pipe. ...

Reinforcement of mortar lining for steel pipe. . .

Concrete masonry around steel pipe

Portland cement

$31.00 Un. ft.
36.00 " "

40.00 '• "

43.00 •' ••

47.00 " "

60.00 " "
2.50 '• "
.02 sq. ft.
6.00 cu. yd.
1.75 bbL

t29.00 Un. ft.
33.00 •• "
88.00 •' "

46.00 " •'

50.00 " "
3.00 •• '•
3.50 " "
.02 sq. ft.
5.25 cu. yd.
1.60 bbl.

Average of
AUBids.

(5 bidder:)

$35.50 Un. ft.

41.20 " "

44.80 " "

48.80 " "

66.80 " "

60.60 " "
4.90 " "

.03 0Q. ft.
6.86 cu. yd.
1.74 bbL

i8 bidden.)

$33.26 Un. ft.

37.37V« " "

42.76 " ••

48.76 " "

63.62Vi " "
3.56»/« " "
4.04»/i *' "
.02«/4 sq. ft.
6.78 ou. yd.
1.72t/s bbL

constructing the Catskill Water. Works. Engineering operations
are being directed by J. Waldo Smith, chief engineer; Robert
Ridgway was department engineer, Northern Aqueduct depart-
ment, until January 14, 1912; Ralph N. Wheeler was appointed
department engineer of that department, February 1, 1912;
Frank E. Winsor is department engineer. Southern Aqueduct
department. In immediate charge of the construction of the
siphons are division engineers John P. Hogan, Alexander Thom-
son, Jr,, and George P. Wood (Northern); George G. Honness,

Digitized by

Google

Flinn on Protection of Steel Pipe with Cement. 443

Emest W. Clarke, and Charles E. Wells (Southern). The draw-
ings and specifications were prepared and many of the preliminary
investigations conducted by Senior Designing Engineer Thomas
H. Wiggin, and Engineer Inspector Ernst F. Jonson has had
charge of inspection of cement, of steel at the rolling mills and
of pipe at the shops, all under the immediate supervision of the
writer. The chief engineer and his personal assistant, Depart-
ment Engineer Thaddeus Merriman, made many examinations
of existing steel pipes which furnished the reasons for seeking a
better protection than the usual coatings.

All drawings reproduced as illustrations are for the 9-ft.
6-in. pipes. Corresponding drawings for the other sizes are
similar; likewise standard dimensions for rivet and joint details.

Digitized by

Google

A FIREPROOF SCHOOL OF CONCRETE.
By Theodore H. Skinner.*

Early in the winter of 1910-11 the writer was intrusted by
the Trustees of Union School District No. 12, Town of Vernon
and City of Oneida, N. Y., with the task of building a new four-
classroom school house which should, in addition to complying
with all the regulations of the New York State Department of
Education, be as nearly fireproof as possible, keep within an
appropriation of $17,000, be so arranged as to appear symmetrical
and complete, while in reality be only one-half of an eight-room
building ultimately desired. The school house shown by the
accompanying illustration is the answer to the many problems
involved in carrying out the task.

The general plans were gone over with the state inspectors
for the purpose of securing informal approval of same before
detailed drawings and specifications were completed. These
plans were then submitted to several parties with requests for
estimates and sketches showing how they would build the frame-
work of floors, roof and enclosing walls. These parties repre-
sented field-cast reinforced concrete, light steel frame with metal
lumber and metal lath stuccoed and factory-made reinforced
concrete.

The sketches received were carefully studied and general
drawings made which would be possible to follow should either
system be selected. Tenders were then invited from a number of
general contractors. The specifications provided that the bidders
might use either one of the three systems proposed and asked
them to name in their bid what would be the difference in cost of
the building if erected by them under the various systems. The
bids ranged from $16,298.00 to $24,500.00 for the bare building
"^dthout plumbing or heating and nothing to spare for moving
furniture, etc. The local contractor, who proved to be lowest
bidder, then opened his estimate books for inspection and it was
found that by combining his own figures for portions of the work

♦ Architect, Oneida, N. Y.

(444)

Digitized by

Google

Skinner on Fireproof School of Concrete.

445

with the figures named by the party bidding on factory made
reinforced concrete that it would be possible to carry out the gen-
eral plans and keep within the appropriation. Accordingly
detailed specifications and drawings were made up for the frame-
work of the building and the general contract let.

The building (Figs. 1 and 2) consists of a rectangular section
25 X 92 ft. running north and south, containing four classrooms
each 24 x 32 ft., four coatrooms 7 x 24 ft., stairway 11 ft. wide,
two playrooms in the basement 24 x 32 ft., separate toilet rooms

FIG. 1. — KENWOOD SCHOOL, KENWOOD, N. Y.

each 7 X 24 ft. for boys and girls and a teachers' room 11 x 13 ft.
with private toilet on the landing over the first- and second-story
stairway. An extension across the front 13 x 40 ft. contains fur-
nace room in the basement, and is entirely corridor on the two
upper floors. There are two minor further extensions forming
vestibules at front and rear.

The plans (Fig. 3) provide for two additional classrooms at
both the north and south ends of the present rectangular section
to be reached by extensions of the present corridors across the
west or blank sides of the present classrooms. Additional stair-

Digitized by

Google

446 Skinner on Fireproof School of Concrete.

{DiAACM^UX ^UHj-

-Lb li IJ- - r^ Z

CCCDDDDDD
inGUDnDDDD

Tim* r^t^/a^TKm .

FIG. 2. — GROUND AND FIR8T FLOOR PLANS, KENWOOD SCHOOL.

Digitized by

Google

Skinner on Fireproof School of Concrete. 447

ways will be provided in the extended corridors and also coat-
rooms for the new classrooms. When these extensions have been
made the west or street face of the building which now presents
large blank panels will be symmetrical on the center line of the
building and when finished will have large windows indicating
the lines of the stairways.

The present classrooms have southeastern exposure with
unilateral lighting only. The coatrooms are accessible only from
the classrooms. Each classroom is provided with a built-in
cabinet of four drawers and two cupboards for the teachers' use
and has blackboards 4 ft. high entirely around the walls except-
ing for the window spaces. The basement rooms are 10 ft. in the
clear, the classrooms 12 ft. The plumbing fixtures are all of
colonial ware and of the latest pattern, two drinking fountains
are arranged in the corridors, one in each story. Special anti-
panic exit bolts are arranged on the outside doors, which open
out.

The details provided for a steel frame for the classroom
section, of 12 columns connected by horizontal steel girders in the
plane of the walls at several points. The girders carried 2 floors
or 2 sets of separately molded and cast reinforced concrete joists,
spaced about 4 ft. on centers which rested on and were anchored
to them. The joists carried in turn a series of ribbed reinforced
concrete slabs separately molded and cast in the factory (Fig. 4).
The roof was constructed in same maimer as the floors, the only
difference being that the joists were not of uniform section
throughout being severally graded or warped so as to give the
roof slabs resting on them the proper pitch to throw the water to
desired points. The floors and roofs of corridors and entrance
porches were built also of separately molded members supported
by bearing walls of masonry. The type of imit members employed
is shown in Fig. 5.

A short description of the process of manufacturing separately
molded reinforced concrete members may be of interest. A
frame was made up of the steel rods necessary to reinforce the
concrete, carefully designed to take care of all the tensile stresses
which might be developed in the member when finished, set in
place and loaded, and also all shear which the concrete would not
take care of. Longitudinal tension rods were carefully bent to

Digitized by

Google

448 Skinner on Fibepboof School of Concrete.

-v{^

Ol "^

trW

M
OQ

Digitized by

Google

Skinner on Fireproof School of Concrete.

449

the desired form and united into rigid frames by means of vertical
loops and small rods womid around the larger ones which also
take care of the shear. Some smaller longitudinal rods were
built into the frames at their tops to take care of any excessive
compression or any tension which might be caused by negative
bending moments produced in handling the beams. Special loops
were attached to the tension rods at their quarter-points extend-
ing up above the top rods in the frames and above the finished
concrete at the top by which the beams were lifted, the strain all
coming on the reinforcement.

PIO. 4. — REINFORCED CONCRETE SLABS IN YARD OP CONTRACTOR.

Sand molds were prepared on large casting floors of the
shape desired for the finished beams and in these molds the unit
reinforcing frames were suspended in proper place, and then
liquid concrete was poured into the molds and thoroughly worked
in and around the reinforcement. The molds were filled to the
top and struck off with the straight edge. A mixture of 1:2:4
concrete was used, the largest aggregate passing a 1-in. screen.

The beams were allowed to remain in the sand 7 days after
which they were lifted by the loops, carried by a traveling crane
into an open yard and stored until ready for shipment.

Digitized by

Google

450

Skinner on Fireproof School of Concrete.

The manufacture of the floor panels would not have been
possible without a vibrating machine. A frame of small rods to
act as reinforcing for the webs was first made, this was then
covered with wire reinforcement well wired to the rods. The
frame was then placed in molds on vibrating machines; the con-
crete poured and vibrated.

The beams were about 10 in. wide, 18 to 24 in. deep, of Tee
section with the top edge rebated on each side to receive the
floor slabs. All joists were approximately 25 ft. long. Floor
slabs were about 4 ft. square webbed or thickened around the
edges and once across the middle to 3 in. thick, the centers were
thin panels only IJ in. thick. Joists and slabs were designed to

FIG. 5. — GENERAL TYPE OF UNIT CONCRETE MEMBERS.

carry a live load of 100 lb. per sq. ft. in addition to their own
weight, with a factor of safety of 4.

The footings for the 12 columns were isolated, 7 ft. square
and 16 in. thick. The basement walls between the columns were
made of 12-in. hollow terra cotta tile carried by independent
footings. The tiles were laid up in cement mortar, plastered with
cement mortar on the outside and coated with asphalt up to
grade. The exterior walls between columns above the basement,
being curtain walls only, were laid up of 8-in. hollow terra cotta
tile and were stuccoed on the outside and plastered on the inside.
The columns and girders were covered with metal lath and con-
creted in solidly.

Digitized by

Google

Skinner on Fireproof School of Concrete. 451

The cornices were supported by a cast stone bed molding,
ornamented with egg dart topped out with a 12-in. tile covered
with stucco. Some small brick inserts were made just below the
cornice to relieve the absolute monotony of the color of the stucco
and shallow lines were drawn in the stucco facia and elsewhere
while it was still soft to form panels and accents. With these
exceptions the exterior is a plain gray floated-finish cement
stucco. It will be given a brush coat of some waterproof cement
stain or finish before another winter to fill up minor checks and
prevent water and frost damage.

All exterior windows were combined frame and sash of metal,
glazed with clear double-thick glass in the lower panels and with
rippled glass in the upper. Each sash was provided with 2 ven-
tilating sections.

The roof is a flat concrete slab pitching only enough for
drainage from the level verge to two outlets near the center line.
This was covered with 5-ply slag and composition roofing and is
not visible from the ground.

The interior finish was the simplest possible. The masonry
walls were plastered with two coats of mortar and left under the
float, then painted 4 ft. high with oil paint two coats, and above
with water color two coats. No wood trim was used around
either doors or windows, but plank jambs of wood were used in
the doorways to which the doors were hung by pivot hinges at
top and bottom. No wood door sills were used, in short the only
wood in the entire building being the door jambs and doors, the
molding at the top and bottom of blackboards, cleats for coat
hooks in the wardrobes and panels of wood flooring in the center
of each classroom.

These panels, 4 in number, were laid of one thickness maple
floor, I X 2J in. face over 2 x 4 in. hemlock sleepers, on top of the
concrete construction previously described, were oiled on the
under side before being laid and given a coat of oil immediately
after laying to prevent their absorbing water from the composition
borders, subsequently laid against them. The balance of the
classroom floors were finished on top of the cement slabs with
concrete surfaced with colored material to match the wood center.
These borders merged into a sanitary or cove base 4 in. high
everywhere. Corridors, stairways and wardrobes all have a

Digitized by

Google

452

Skinner on Fireproof School of Concrete.

granolithic cement coat over the structural concrete slabs, this
finish also merging into a sanitary base 4 in. high. There was no
projection to these sanitary bases at the top, the same being
finished flush with the finish plaster.

The four classroom ceilings were plastered upon expanded
metal lath carried by angle-iron furrings which in turn were

FIO. 6. — CEILING OF PLAYROOM, KENWOOD SCHOOL.

attached to numerous iron lugs left projecting from the cast con-
crete floor joists for this purpose. These furred ceilings were
used for a double purpose, first to make a ceiling free from all
shadows, and, second, to provide a deadening air space so that the
noise of walking on the top of the concrete construction should
not be heard in the room below. The basement, corridor and

Digitized by

Google

Skinner on Fireproof School of Concrete. 453

stairway ceilings were left unfurred, with the cast joists and panel
slab construction showing, as may be seen in Fig. 6, all painted
with water color paint presenting a very attractive appearance.

The contract, let May 15, 1911, provided for the completion
of all work by September 15, but delays were experienced from
the start, first in getting the necessary steel fabricated and later
in securing competent labor for erecting the steel and laying the
heavy concrete floor and roof joists. It was not until November
26, or a few days over six months from the start, that the work
was completed.

The joists and slabs were loaded on cars at the factory and
shipped by local freight 28 miles; were unloaded by means of a
small traveling crane on to farm wagons and then hauled a little
better than a mUe to the job. Here they were unloaded and
swxmg into place by seemingly inadequate apparatus, all at less
expense than the contractor figured he could erect forms and cast
the same number of members in situ. The joists and slabs stood
shipping remarkably Well and reached the job in good condition
in spite of the rough handling given by the local train, on which
the cars are shunted back and forth at every station a great num-
ber of times. Only one joist and four slabs required replacing as
they were too green when loaded, 30 days* seasoning should be
allowed before shipping.

The straightforwardness of the various operations at the
building appealed both to the general contractor and to the
architect as it offered opportunity to inspect each member before
erection and the job was not littered up nor complicated with the
forest of supports necessary for field concrete construction. No
unexpected difficulties were encountered and no members failed
to fit their respective locations accurately, and everyone con-
nected with the work would be satisfied to repeat the operation
again, with the exception possibly of the maker of the cast work,
who might wish to add slightly to the original allowance for
handling the cast members at the building.

The estimates of cost for the various types of construction
not used were perhaps confidential and will not be given in detail ;
it is sufficient to say that they were higher than those obtained
for the construction adopted.

Digitized by

Google

454 Skinner on Fireproof School of Concrete.

The contracts for the building as erected were as follows:

Steel frame $1,500.00

Structural concrete floors, roof and stairs 3,000 . 00

Suspended ceilings, unplastered 225 . 00

Granolithic in halls and coatrooms 196 . 00

Colored surfacing and sanitary base in classroom 379.00

Iron stair rails 200.00

General contract for balance of work and materials, including ex-
cavation, grading, carpentry, painting, etc 9,545.00

$15,045.00

Plumbing 767.00

Heating 825.00

Hardware 225.00

Total $16,862.00

In addition to the above a flowing spring was encountered in
the excavation which gave standing water over part of the base-
ment and made it desirable to underdrain same and waterproof
basement walls; it also necessitated a change in the plans of the
cold-air boxes from beneath the basement floor to overhead.
These two items cost respectively $200.00 and $227.97. Some
additional fireproofing of steel to that provided under the con-
tract was thought desirable and added at a further expense of
$400.00. Cement platforms or walks outside at the entrance
cost $38.00, making a total of extras on the work over and above
the first contracts of $863.97 and bringing the total cost of the
building as delivered by the contractor to the Trustees, $17,527.97.

Not coimting the area of vestibules, this building has an
area of 2820 sq. ft. on the ground, is 40 ft. from bottom of base-
ment floor to top of roof and contains 112,800 cu. ft. Its cost
per cu. ft. is 15.5 cts., per classroom $4381.99 and per pupil as
$97.37.

Digitized by

Google

THE PRESENT STATUS OF UNIT CONCRETE
CONSTRUCTION.

By James L. Darnell.*

To put it briefly it might be said that the present status of
Unit Construction is one of progression. Every day new things
are developing about this very logical and rational method of
constructing and erecting concrete structiures. Within the past
two years there has been more progress in both the design and
in the field work, that is, the construction and erection of sep-
arately molded structures, than in all the time previous.

In heavier structures, such as bridges and viaducts, the
engineering forces of the Chicago, Burlington and Quincy Railway
Company under the direction of Mr. C. H. Cartlidge, Bridge
Engineer, and of Mr. George E. Tebbetts, now bridge engineer
for the Kansas City Terminal Railway Company, have done
much valuable work and they have pioneered the way for others
less advancced. The Chicago, Milwaukee and St. Paul Railwaj'^
Company under Mr. C. F. Loweth, Chief Engineer, has also
done some work in this line, but not to the same extent. Mr.
R. E. Gaut, while bridge engineer of the Illinois Central Railway
Company, also used this system in the design of highway struc-
tures in connection with track elevation work in Chicago. All
of this work was successfully done and showed pronounced
advantages over ordinary methods.

If this is the "Concrete Age" as most of us fondly believe,
it will certainly come to pass that the railroads particularly will
have to adopt unit construction methods for their concrete struc-
tures because it lends itself with peculiar fitness to railway work
of all kinds and shows such marked economies in both time and
money, that ordinary or monolithic construction is out of the
question. This is shown in the present track work of the Kansas
City Terminal Railway Company, where Mr. Tebbetts has
adopted this form of construction for every structure along the

* Manager. Kansas City Unit Construction Company, Kansas City. Mo.

(455)

Digitized by

Google

456 Darnell on Unit Concrete Construction.

line except those where surrounding conditions are such that the
use of some other type was imperative, as the McGee Street
Viaduct, where the 124-ft. span required a concrete encased
steel girder.

In building construction the progress has been more marked
perhaps than in bridges and viaducts. The first recorded example
of a building of any magnitude built of separately molded units
is that of the two kiln houses for the Edison Portland Cement
Company at New Village, N. J., early in 1907.* These two
buildings were simple one story sheds of the plainest possible
type. No attempt was made to elaborate or to refine the design,
as nothing more was necessary than a simple assemblance of
columns, roof beams and roof slabs. In these two sheds however,
the soundness of the principle of separately molded units was
demonstrated.

At about this same time the idea of building construction
with separately molded units seems to have been taken up inde-
pendently in other sections of the country. In the east Mr.
E. L. Ransome was developing a system which he is employing
successfully up to the present time. Perhaps the most notable
example of the Ransome type of Unit Construction is the four
story building 60 ft. x 200 ft. built for the United Shoe Machin-
ery Company at Beverly, Mass. This building has successfully
met all its requirements and is in every respect equal to and in
some respects superior to similar buildings of ordinary mono-
lithic construction.

Mr. Charles D. Watson of Syracuse, N. Y., has also done
considerable work along this line, having successfully constructed
several separately molded buildings, f

In St. Louis, at about this same time, Mr. Albert J. Meier
in conjunction with Mr. John E. Conzelman, at that time engi-
neer for the Corrugated Bar Company, and Mr. C. D. Morely
a contractor, were working to perfect a Unit system independently
and without knowledge of the work that was being done by others.
From the beginning there were two very apparent difficulties to
be met and surmounted. The first one was the difficulty in devis-
ing or designing connections between the separately molded units

♦ Seo Proceedings, Vol. IV. 1908. p. 48— Ed.

t See ProceediTHjB, Vol. IV, 1908. p. 97; Vol. VI. 1909, p. 391.

Digitized by

Google

Darnell on Unit Concrete Construction.

457

which would be as strong as the units themselves. Second, there
were well taken objections on the part of many people to sep-
arately molded buildings of concrete on account of the pure and
monotonous ugliness of such buildings.

The first difficulty, that of the connections, has been most
successfully overcome by the use of interlocking and overlapping
reinforcement, broad bearings and improved methods of grouting
the connections. The second or aesthetic objection is being
gradually overcome both by improved and advanced design and
by a process of evolution, in which the engineer in designing is

FIG. 1. INTERIOR OF WAREHOUSE, NATIONAL LEAD COMPANY, KANSAS CITY, MO.

slowly but surely drawing away from the architectural faults of
its monolithic predecessor. The progress is best shown in the
illustrations which follow.

The first building constructed with separately molded
units by the Unit Construction Company of St. Louis, Mo.,
was a warehouse for the National Lead Company, erected on
West Thirteenth Street, in Kansas City, Fig. 1. This build-
ing was built under great difficulties in a very restricted
space. It was attempted to mold the units within the building
lines, which proved to be very expensive by reason of the neces-

Digitized by

Google

458

Darnell on Unit Concrete Construction.

sary rehandling of the unite from time to time. Because of the
unsightly appearance which would have been presented by the
use of concrete unite in the front of the building, the National
Lead Company at that time wanted a brick face. The rear of
the building, however, was built of concrete unite and ite appear-
ance in a measure justifies the objections of the owners on the
score of unsightliness. The different pointe between the unite
are very cleariy indicated and while the wall on the whole does
not present a very beautiful appearance, it is after the lapse of
some four years, perfectly serviceable, thoroughly substantial
and water-tight. We are certainly proud of this, our first build-

FIG. 2. CASTING YARD FOR CONCRETE UNITS.

ing, because it is still standing up, apparently more substantial
than when built and because it has been so thoroughly satis-
factory to the owners.

This first building was so successful that the company was
enabled, in competition with various other contractors, to secure
a contract for a very much more extensive construction for the
National Lead Company in St. Louis. Fig. 2 is a general view
of the casting yard laid out for this construction. Fortunately
there was in this case plenty of room to spread out in, conse-
quently it was not necessary to cast the unite on the ground
within the building lines. An extensive plot of ground was avail-

Digitized by

Google

Darnell on Unit Concrete Construction. 459

FIG. 3. — SETTING A WALL SLAB.

FIG. 4. — ERECTION OF CORRODING STACK HOUSE, ST. LOUIS, MO.

Digitized by

Google

460

Darnell on Unit Concrete Construction.

able adjacent to the site of the building and thereon was erected
the construction plant. Concrete materials were elevated in a
belt and bucket conveyer from cars to bins in the top of the
tower, seen in the left center of the picture. Immediately below
these bins was the concrete mixer, which discharged by gravity
into the cars on the oval elevated track. Movable spouts con-
ducted the concrete from this track level to the molds for the
units which were laid horizontally in the casting yard. A travel-
ing derrick was erected to run between the elevated tracks and
was used to handle the completed units from the yard onto flat
cars for conveyance to the building site.

FIG. 5. — ROOF SHOWING CONCRETE SKYLIGHTS AND LOUVRES.

Fig. 3 is a view of the building after construction had some-
what further progressed. This shows many of the columns in
place and the workmen setting a partition slab. It may be seen
that the foundations were run in monolithically with slots left
for the reception of the columns. A key was cast in each slot
to engage a corresponding slot cast in the columns, both of which
may be plainly seen in the picture. Fig. 4 shows a view of the
structtire when about two-thirds complete. This view illustrates
another advantage of unit construction in that as soon as any
part of the building is in place it is ready for service. This was
peculiarly demonstrated in this particular building. To make
room for this structure it was necessary to dismantle and tear

Digitized by

Google

Darnell on Unit Concrete Construction.

461

down the old wooden stack building and necessarily the machine
capacity was seriously impaired. It proved to be of great
advantage that after four or five bays were finished, they were
turned over for service. As it happened the compan}*- was
making white lead in more than three-fourths of the building by
the time it was finally completed. Fig. 5 is a view of the roof
of this same building, showing the sky lights and louvres, and
illustrates the roof construction which was of the regulation felt,
tar and gravel type. This building when complete was 750 ft.
long, 105 ft. wide and about 45 ft. high.

Fig. 6 is a group of three buildings built for the National
Lead Company at their St. Louis plant, contracts for which were

FIG. 6. — lU 1LDINCJ8 OF NATIONAL LEAD COMPANY, ST. LOUIS, MO.

secured after the stack house had progressed pretty well along
toward completion. This group consists of an office and welfare
building on the right, an oil house in the center, and a stable
and garage on the left of the picture. In this group is seen
something of the architectural progress which was referred to
earlier in this paper. They certainly present a not unpleasant
appearance and we think they appear favorably with some build-
ings of similar type in any material.

Figs. 7 and 8 represent the most pretentious structure which
has engaged the attention of the Unit Construction Company
up to the present time. This is an oxide mill also for the National
Lead Company, built in St. Louis, which provided for floor loads
running from 500 lb. on the first floor to 250 lb. on the top floor.

Digitized by

Google

462 Darnell on Unit Concrete Construction.

FIG. 7. — OXIDE MILL IN COURSE OF ERECTION, FIVE STORIES.

fc ,- _J^

FIG. 8. — OXIDE MILL BUILDING COMPLETED.

Digitized by

Google

Darnell on Unit Concrete Construction.

463

A study of these views will give to the observer an excellent
idea of the construction methods employed as nearly all of the
details may be observed.

Fig. 9 shows a corroding stack building erected for the
National Lead Company at New Kensington, Pa. They are
interesting in that they show architectural progress as well as
improvement, both in design and workmanship.

Last year the first contract was taken for a 50,000 bushel

FIG. 9. — CORRODING STACK BUILDING, NATIONAL LEAD COMPANY,
NEW KENSINGTON, PA.

grain elevator, Fig. 10, for the Highland Milling Company at
Highland, 111. In this case each slab unit had a column cast on
the end of it and the whole was held together by big rods running
entirely through the building with large cast washers on the
outside. This elevator had nine bins in it; eight were storage
bins and the ninth carried the elevating machinery. This struc-
ture has been in service for more than a year and the owners of
the Highland Milling Company are very enthusiastic advocates
of Unit concrete construction because during a season in which

Digitized by

Google

464

Darnell on Unit Concrete Construction.

it was very hard owing to weather conditions to successfully
store grain, in this elevator not a bushel was lost.

A building under construction for the Ohio Cultivator Com-
pany at Bellevue, Ohio is shown in Fig. 11. This building is
200 ft. square, three stories high with a high basement.

FIG. 10. — GRAIN ELEVATOR, HIGHLAND MILLING COMPANY, HIGHLAND, ILL.

As to the use of Unit construction in railway work, Fig. 12
shows a reinforced concrete crossing at Sangoman Street in
Chicago, 111. The work of the Kansas City Terminal Railway
Company where concrete units are being used for all the via-
ducts and subways required in the Terminal Company's improve-

Digitized by

Google

Darnell on Unit Concrete Construction.

465

ments, which are now going on, includes some forty structures
in all. Fig. 13 shows a unit bridge; Fig. 14 column molds as
well as some of the finished columns; Fig. 15 shows an outer
deck slab which carries the coping and panel fascia.

The pictures show for themselves the present status of Unit
Construction both as applied to buildings and to bridges and
kindred structures.

A word as to the '*why" of unit construction. At first

FIG. 11. — BUILDING FOR OHIO CULTIVATOR COMPANY, BELLEVUE, OHIO.

glance, to an engineer, the system looks novel and interesting,
but of narrow application and of little practical value or utility.
Invariably, however, after thorough study the engineer becomes
more and more convinced of the practicability of the system and
of its broad and diverse possibilities. In fact those who are con-
nected with its application cannot find the time to go into the
many virgin fields of construction effort, but up to the present
have been compelled to devote all their time and energies to
the development of the single field of mill and warehouse con-

Digitized by

Google

466

Darnell on Unit Concrete Construction.

struction with short excursions into the grain elevator and rail-
way fields.

For unit construction are claimed all the advantages of

FIG. 12. — CONCRETE BRIDGE CROSSING, SANGOMAN STREET, CHICAGO, ILL.

FIG. 13. — RAILROAD BRIDGE OF UNIT CONCRETE COLUMNS, BEAMS AND SLABS.

permanency, strength and ultimate economy to which structures
of reinforced concrete are justly entitled, without any of the
disadvantages and uncertainties which in the structure built by

Digitized by

Google

Darnell on Unit Concrete Construction.

467

ordinary methods are so often attended by unfortunate results —
accidents which are always serious, sometimes fatal.

By this method the designer is sure that no expansion joints

FIG. 14. — UNIT CONCRETE COLUMNS AND MOLDS.

FIO. 15. — UNIT CONCRETE DECK SLAB WITH COPING.

will open up in unexpected places because of improper deposition
of concrete or other carelessness of the workmen. Each unit
member of the structure is built on the ground under practically

Digitized by

Google

468 Darnell on Unit Concrete Construction.

factory conditions. Eflfectual inspection is assured in all respects.
The reinforcement is placed where the engineer designed it to
be; each unit is a true monolith without seam or flaw; each
unit is properly cured and is prepared in the building; the pos-
sibility of getting an imperfect column, beam, girder or slab into
a structure is entirely eliminated. In short, the manufacturing
conditions of the units make for a maximum of efficiency in all
departments with the chance for uncertainties reduced to prac-
tically a negligible quantity. In erection the same condition
prevails as the mechanical handling of the units is exactly the
same as in the handling of steel and with properly designed units
it is practically impossible to go wrong on the connections.

Unit Construction today is just beginning to come into its
own. Its application is wide and the genius of the American
engineer will soon lead to the general adoption of a system which
combines in itself the maximum of efficiency, economy and
permanency and minimizes the uncertainties contingent on the
vagaries of the American workingman — ^who is principally a
foreigner.

Digitized by

Google

DISCUSSION.

Mr. Rudolph P. Miller. — Mr. Chairman, I would like to Mr.Muier.
inquire if the joints must not be filled up on the work and whether
any attempt is made to carry the reinforcement from one member
to an adjoining member so as to effect continuity?

Mr. James L. Darnell. — ^The joints and connections are Mr. Dameu.
all grouted in. The reinforcement overlaps and interlocks, i.e.
goes through the connection which results in practically a mono-
lith. In fact, extensive laboratory tests show that the connec*
tions are stronger than the units themselves. It has been demon-
strated in buildings actually constructed and in service that the
connections are the strongest part of the structiu'e.

Mr. John E. Conzelman. — ^The question just asked would Mr. conzciman.
indicate that there was some doubt in the speaker's mind as
to the stability or rigidity of the buildings and I want to say
that the buildings we have constructed seem to be just as rigid
and stable as similar buildings constructed in the ordinary
manner. In fact after the joints are poured a imit constructed
building has the advantages of continuous action and rigidity
characteristic of buildings made by the ordinary or monolithic
method. It may be interesting to know that the thin walls used
on these buildings have proven satisfactory. We have con-
structed seven or eight buildings with three and four inch walls
which have passed through two winters with no complaints from
the owners.

Mr. E. J. Moore. — The question of shrinkage cracks in Mr. Moore,
floors seems to be important. The speaker referred to the absence
of cracks in the floors. Cracks from temperature stresses will
no doubt occur at the joints where the different sections are put
together and that would seem to make an added problem of
waterproofing the joints.

Mr. Conzelman. — Mr. Darnell stated that cracks often Mr. conzeiman.
occurred in concrete structures built in the usual way and that
these cracks did not at all times select their location with due
regard for the appearance of the structure or the feeling of the
builder; in fact these cracks often appear at imfortunate places.

(469)

Digitized by

Google

470 Discussion on Unit Concrete Construction.

Mr. ConzeinuiA. It is unusual to 866 a long concr6t6 retaining wall or building in
which there are not some cracks. These cracks (neglecting settle-
ment or other structural causes) are due to two causes, shrinkage
of the concrete as it dries out and temperature changes, acting
separately or in combination.

A great many of the cracks are due to shrinkage stresses
but in unit construction these stresses are practically elminated.
This is due to the fact that each vmit has hardened before it is
incorporated into the building. The unit itself may have some
slight internal stresses due to shrinkage, but these are not cumula-
tive and do not affect the complete structure. Each unit takes
care of itself.

Temperature stresses cannot be eliminated, although tem-
perature changes undoubtedly have less effect on unit construc-
tion than on monolothic, on the same principle that a brick wall
(which is constructed of small units) will generally show fewer
cracks than a concrete wall. We have observed this action care-
fully and have had very little difficulty, if any, from temperature
effects and account for it by the fact that during the construction
of the building and before the connections are grouted, the imits
attain an average temperature.

Mr. Moore. Mr. Moore. — I take it therefore that it is necessary to

waterproof most of this filling. Joints in factory floors are more
especially referred to as these are washed down and there is
objection to having the water go through the cracks.

Mr. conxeinum. Mr. Conzelman. — ^Thc construction shown is absolutely

waterproof so far as we know. A unit weighing 4 or 5 tons
when set on. a mortar bed will compress it to such an extent
that it is practically watertight. One advantage of unit con-
struction is that we know just where the joints are and as leak-
age is necessarily confined to the joints proper provision can be
made.

Mr. Kinney. Mr. William M. Kinney. — It would secm there would

certainly be some little unevenness, which is one of the greatest
troubles of the concrete floor and one of the things we are trying
to eUminate in floors that have to be trucked over. The trouble
occurs where the joints are made in a floor and the question
arises whether in unit construction any {protection is afforded
over the joints.

Digitized by

Google

Discussion on Unit Concrete Construction. 471

Mr. Darnell. — ^There is no apparent joint at the edge of Mr. Dameu.
the floor slabs. The floor slab and beam are cast in one and
the edges of the slabs are made so that the space between is in
the form of a wedge, which is filled up level with mortar so that
practically there is no joint left at all and the floor is perfectly
smooth. That is perhaps one of the advantages of this form of
construction. The edge of the floor slab is built to exact dimen-
sions and if any vmevenness develops it is very easy in setting
the slab on the girder to level it up smoothly, which is done.
If the levelling is carefully done and these wedge shaped joints
filled up with grout, it makes the floor, to all intents and pur-
poses so far as the surface is concerned, perfectly smooth. There
has been no trouble with uneven joints in trucking over floors
in any of the buildings. The National Lead Company's buildings
in St. Louis, is a group in which trucking is constantly going on
and there has never been a complaint at any time.

Mr. Allen Brett. — ^The tests reported by the Committee Mr. Brett,
on Reinforced Concrete and Building Laws, in the case of the
beam and slab show that the concrete in the slab acts in com-
pression practically all the way across. This girder used in unit
construction is a broad inverted T, practically, and the slabs are
inverted boxes, the edges of which rest on the flange of the T.
Is there any compression acting with the girder in the slab?

Mr. Conzelman. — ^The girders are made with ledges or Mr. conzeinum.
shelves on each side upon which rest the floor slabs, usually made
to resemble the cover of a large box. They consist essentially
of a thin plate or slab carried by beams on each side, these beams
returning arovmd the ends of the slab; the arrangement gives the
slabs a continuous or uniform bearing on the girder ledge. The
thickness of the slab or plate may vary from IJ to 5 or 6 in.
depending on the load to be carried and the span between beams.
When the slabs are placed the top of the slab is higher than the
top of the girder, an amovmt equal to the thickness of the slab,
and the steel projecting from the slabs on each side of the girder
interlock in the space so formed; the stirrups from the girder
also extend into this space. The joints between the girder and
slabs are then grouted and the space filled with a rich concrete.
After the grout has hardened these become for all practical pur-
poses T beam sections.

Digitized by

Google

472 DiscuBsioN ON Unit Concrete Construction.

Mr. cm&zeiiiuui. This T-beam action has been demonstrated in the laboratory

on large beams by measuring the deformation in the concrete
directly over the girder and in the adjoining floor slabs and also
the deformation of the reinfprcement, by means of an extenso-
meter similar to that which has been described by the Com-
mittee on Reinforced Concrete.

Mr. H. p. Green. Mr. HERBERT P. Green. — What provision is made to take

up the shear between the columns other than the brackets on the
columns?

Mr. Conzeinum. Mr. Conzelman. — The brackets are designed to take the

entire load from the girder; the brackets are designed for ver-
tical shear and bending moment. The size of the top of the
bracket is determined by the bearing area required to properly
distribute the load from the girder to the bracket.

Digitized by

Google

REPORT OF COMMITTEE ON SPECIFICATIONS AND
METHODS OF TESTS FOR CONCRETE MATERIALS.

In this first report of the Committee no definite recommenda-
tions are presented for specific tests or methods of making tests
of aggregate, except that the Committee reconmiends for a practi-
cal test of sand a determination of the strength of mortar made
with it. The report, therefore, is in the nature of a tentative
discussion of various tests and of methods that have been used
in diflFerent laboratories. This progress report should be followed
during the coming year by further investigations leading to more
definite reconmiendations.

The Committee request that information on methods of mak-
ing tests of aggregates and results obtained by such methods be
forwarded to the chairman of the committee. Laboratories that
are in a position to assist the Committee or imdertake research
work in the line of concrete aggregates are also asked to corre-
spond with the Committee.

Importance of Testing the Aggregates.

The selection of aggregates for concrete, especially the selec-
tion of sand or fine aggregate, is of as great importance as the
selection of the cement. So evident is this to the engineer who has
had experience both in practical construction and in laboratory
tests, that it is almost inconceivable that so much important
work should be undertaken and carried through without testing
the sand. Frequently every carload of cement is carefully sampled,
but no test whatever, except by inspection, is made of the equally
important ingredient, sand. When tests are made, they frequently
are confined to mechanical analysis or granulometric composition.
While this is a valuable test for comparing the qualities of different
fine aggregates that other tests have shown to contain no impuri-
ties, it does not show up some of the worst defects that occur
occasionally. It therefore cannot be relied upon alone.

The impossibility of determining the true quality of a siand
or other fine aggregates by mere inspection cannot be emphasized
too strongly.

(473)

Digitized by

Google

474 Report on Tests for Concrete Materials.

Sampling Aooreoates.

Samples must be taken in such a way as to obtain a fair
average of the material to be tested.

The size of the sample depends upon the character of the
aggregate and the nature of the test. One should err on the side
of getting too large a sample rather than one that is too small.
For tensile or compressive tests of mortar made with fine aggre-
gate a sample not less than 20 lbs. in weight should be taken in
order to have enough material left over for other laboratory tests
that may be deemed necessary. If practical tests of proportions
with coarse aggregates are to be made, the sample of fine aggregate
should be several times larger than this.

The coarse aggregate sample should be larger than that of
the fine aggregate in order to get fair average of the material,
because the grains are larger and there is more variation in them.
Whatever tests are made must be on a larger scale. For tests
involving both mechanical analysis and volumetric tests of con-
crete mixtures for proportioning the aggregates, at least 200 lbs.
of each coarse aggregate are needed.

Samples should be shipped in a strong box or a bag. It is
advisable also that the natural moisture be retained as far as po"^-
sible, so that the laboratory will receive the material in its natural
condition.

For sub-dividing the sample to obtain the required amount
for each test, different methods are employed in different labora-
tories. One of the common methods is that of quartering.

Quartering. — To quarter a sample of aggregate, it is spread
out on a thoroughly clean floor or table, or else upon a large sheet
of manilla paper. Care is used in spreading to see that particles
of different size are distributed through the mass. The pile is
preferably in the shape of a circular disc. The material in this
shape is divided into four quarters. Two opposite quarters are
removed, taking care to remove all dust. The remaining quarters
are then mixed together. After mixing, the material is spread
out again, as before, and quartered again. This process is followed
until the quantity remaining is of the size required for the ex-
periment.

Digitized by

Google

Report on Tests for Concrete Materials. 475

Tensile or Compressive Test of Mortar.

The Committee recommends the test of the strength of mortar
as that best indicating the quality of the fine aggregate.

To eliminate variations in result, due to the character of the
cement, the difference in laboratory conditions, and the .personal
equation of the operator, the test always must be a comparative
one. For comparison, standard Ottawa sand which is used for
cement tests, and which every well-equipped laboratory should
be provided with, is recommended.

The Joint Committee on Concrete and Reinforced Concrete
in their 1908 report, makes the following recommendation for this
test of strength :

Mortars composed of one part Portland cement and three parts fine aggre-
gate by weight when made into briquets should show a tensile strength of
at least seventy (70) per cent, of the strength of 1 : 3 mortar of the same con-
sistency made with the same cement and standard Ottawa sand.

While this requirement is far in advance of usual practice,
where no laboratory tests are required, it is not so severe as
should now be demanded in the present state of the art of rein-
forced concrete construction. For the present, the Committee
recommends that sand used in reinforced concrete be accepted
only after tensile or compressive tests of 1 : 3 mortar made with
the sand in question, in comparison with similar tests of mortar
of standard sand made up at the same time under the same
conditions.

The Committee is not yet prepared to recommend a fixed
value for the ratio of strength for acceptance. It is suggested
for the present that the ratio be set to suit individual requirements.

To avoid the removal of any coating on the grains which
may affect the strength, bank sand should not be dried before
being made into mortar, but should contain natural moisture.
The percentage of moisture to use for correcting the weights in
measuring the proportions may be determined upon a separate
sample. From 10 to 40 per cent, more water may be required
in mixing bank sands or artificial aggregates than for standard
Ottawa sand to produce the same consistency.

In the mortar tests, enough test pieces should be made to
test at 72 hours, 7 days and 28 days, the first 24 hours all being
stored in moist air, maintained at a temperature of 70 deg. Fahr.,

Digitized by

Google

476 Report on Tests for Concrete Materials.

and the remainder of the period in water at the same tempera-
ture. The 72-hour test is the mast severe, and sand failing
to attain this requirement frequently reaches it at 7 days or at
28 days, and can then be accepte^d. If, however, the 72-hour
briquets break in the clips of the machine, or if the test pieces
at this age show very low strength, say 25 per cent, or less, of
the strength of standard sand mortar, the sand should be considered
dangerous to use on any important work of construction.

Mechanical Analysis.

For proportioning fine and coarse aggregate, the tests of
mechanical analysis are important. Curves of mixtures in dif-
ferent proportions, based on the combined analyses of the cement,
fine aggregate, and coarse aggregate, may be drawn and studied
to obtain the proportions corresponding most nearly to the ideal
requirements. The proportions thus found may then be used
in tests of volume, as referred to below.

Mechanical analysis of fine aggregate is valuable as furnish-
ing an indication of its quality. It is recommended that fineness
requirements be introduced into concrete specifications. This
test cannot be relied upon fully, however, since there may be
impurities in the sand that will make it unfit for use even when
the analysis is satisfactory. The chemical and also the min-
eralogical composition of the sand also may afifect its strength.
Leaving out of consideration, however, the question of impurities,
for which specific tests probably will soon be evolved, the test
of mechanical analysis, or granulometric composition, as it is
sometimes termed, is worthy of much further development.
Laws which govern the effect of the sizes of the particles of the
aggregate upon the resulting mortar or cement are not yet clearly
formulated.

Further studies are necessary for the selection of standard
sieves for use in mechanical analysis. From the report of the
Committee on Reinforced Concrete of this Association,* the fol-
lowing paragraphs are quoted:

19. The relative strength of mortars from different sands is largely
affected by the size of the grains. A coarse sand gives a stronger mortar
than a fine one, and generally a gradation of grains from fine to coarse is

♦ See ProeeedingM, Vol. V, p. 457. — Ed.

Digitized by

Google

Report on Tests for Concrete Materials,

477

advantageous. If a sand is so fine that more than 10 per cent, of the total
dry weight passes a No. 100 sieve, that is, a sieve having 100 meshes to the
linear inch, or if more than 35 per cent, of the total dry weight passes a sieve
having 50 meshes per linear inch, it should be rejected or used with a large
excess of cement.

20. For the purpose of comparing the quality of different sands a test
of the mechanical analysis or granulometric composition is recommended,
although this should not be substituted for the strength test. The percentages
of the total weight passing each sieve should be recorded. For this test the
following sieves are recommended:*

0.250 inch diameter holes, t

No. 8 mesh, holes 0.0955 inch width No. 23 wire.
No. 20 " " 0.0335 " " No. 28 "

No. 50 " " 0.0110 " " No. 35 "

No. 100 " " 0.0055 " " No. 40 "

21. The effect of mechanical analysis or granulometric composition
upon the strength of mortar is illustrated in Appendix. By this table (which
follows) the relative strength of different sands may be approximately estimated.

TssTs BY New York Board of Water Supply of 1:3 Mortar
Made with Sands of Different Mechanical Analysis.

Peroentaces Paadns Sioves.

Tenmle Testa.

CompreasioD Test.

No. 4. 1 No. 8.

No. 50.

No. 100.

7 days.

00 days.

7 days.

90 days.

100 ' 70

213

613

2690

5640

100

263

412

1915

4660

100

177

325

905

2170

100

178

282

1070

1500

100

139

228

905

1130

100

122

170

275

810

100

149

330

490

Void Tests.

Void tests of coarse aggregates frequently are used to de-
termine proportions for concrete. They do not give entirely
correct results, however, unless the tests are made with mixtures

* Sheet brass periorated with round holes passes the material more quickly than square
Round holes corresponding to sieves No. 8, 20 and 50 respectively are approximately
0.13&, 0.060, 0.020 inch diameter.

t A No. 4 sieve, having 4 meshes per linear inch, passes approximately the same sise
gnUns as a sieve with 0.25 in. diameter holes.

Digitized by

Google

478 Report on Tests for Concrete Materials.

of all the ingredients. Usually some of the grains of the fine
aggregate are so coarse as to force apart the grains of the coarse
aggregate. If the sand is fine in proportion to the stone, less
mortar will fill the voids of the stone than if the sand is coarser
and therefore more nearly the size of the stone particles.

Void tests of fine aggregate are also affected by the fact
that the cement forces the grains of sand apart. The voids in
fine aggregate also are affected to a large degree by the percentage
of moisture contained in it. If perfectly dry, a fine aggregate
with grains of uniform size may have nearly the same percentage
of voids as a coarse aggregate of uniform size grains, although
the former will produce a very weak mortar and the latter a
strong one. If the voids were based on the volume of the moist
aggregate, the results would be more normal, but a slight varia-
tion in the percentage of moisture produces such a marked effect
that it is impossible to make true comparisons in this way.

A common method of determining the voids in an aggregate
is to place it in a measure, either loose or compacted as desired,
and measure the quantity of water which can be poured in. The
percentage of voids, either by weight or volume, is thus found
directly. This method with a clean coarse aggregate is fairly
satisfactory. With a fine aggregate, however, air is entrained
and it is almost impossible to obtain correct results.

Another plan sometimes followed is to measure the material ;
place a definite quantity of water in a graduated vessel ; pour the
aggregate into the water; and determine from the graduated
scale the difference in volume of the water before and after add-
ing the aggregate. This difference is the amount of water which
the aggregate displaces. This is substantially a specific gravity
method.

A still simpler plan, if the specific gravity of the aggregate
is known or can be readily determined, is to weigh a given bulk
of the aggregate, loose or compacted as required, correct for
moisture, and compute the voids directly.

Weight.

Weight tests have the same limitation as void tests, since
the weight is affected by the percentage of moisture contained
in the aggregate. The weight also varies directly with the specific

Digitized by

Google

Report on Tests for Concrete Materials. 479

gravity. If the specific gravity of the material is known and the
percentage of moisture is determined, the voids can be computed,
as indicated above.

Volumetric Tests op Mortar of Fine Aggregates.

If there are no organic, or other similar impurities, that ab-
normally reduce the strength, the aggregate producing, with the
cement and mortar, the smallest volume of mortar or concrete
is apt to give the greatest strength.

A volumetric test is better than an ordinary void or a weight
test because the aggregate is mixed with cement and water as
in practice.

The general method employed in making a test of volume
is to mix up the aggregates in the given proportions by weight,
add enough water to produce a consistency slightly softer than
used in tensile tests, and determine the bulk of mortar or con-
crete made with this mixture. Knowing the specific gravity of
all the materials used, the absolute volumes and density can be
computed. The method of making this test is described more
fully in a paper on Concrete Aggregates presented in 1906.*

Volumetric Tests of Concrete Ingredients for
Determining Proportions.

One of the most valuable fimctions for volumetric tests
lies in determining the proportions of concrete. The value of
the test is based on the principle that, with the same proportion
of cement, the mixture which gives the smallest volume, and ia
therefore the densest, usually produces the strongest concrete.
This rule is not strictly true for permeability because the size
of the voids as well as the density influence the permeability.

For determining the density of the concrete, the specific
gravity of each aggregate must be known. The specific gravity
of cement may be assumed as 3.10. An average specific gravity
for bank sand is 2.65.

The process of making volumetric tests of mixtures of coarse
and fine aggregates with cement is similar in principle to the
volumetric test of mortar described. Larger volumes must be

♦ See Proceedirtifa, Vol. II, p. 27.— Ed.

Digitized by

Google

480 Report on Tests for Concrete Materials.

used and it is sometimes easier to fill a measure of given size
and determine the amount left over than to determine the bulk
of the total material. A blank form for use in this test is referred
to below.

Microscopical Examination.

The mineralogical composition of a fine aggregate may fre-
quently be determined approximately by an examination under a
microscope or magnifying glass of high power. Quartz can be
recognized and if dirt surrounds and adheres to the grains it can
be seen.

Chemical Analysis.

The value of chemical analysis is not clearly defined. A
quartz sand in general is better than a natural sand of other
composition, chiefly because it is cleaner. It is claimed by some
that the amount of clay material in the aggregate affects the sand
in other ways besides increasing the amount of fine material.

The effect of colloids and colloidal action remains to be
studied.

The ignition test referred to below is really a chemical test.

Test for Organic Matter.

Experience with defective concrete indicates that the quality
of a sand may be very poor through a minute quantity of organic
or vegetable matter contained in it. The best method of testing
for this and the limitations which must be placed upon the quan-
tity are not yet clearly defined.

A method of test was suggested several years ago by the
Chairman of the Committee and has been used in practice by
him and also, more recently with some modifications, by another
member of the Committee, Mr. Chapman. The methods are
substantially as follows:

Two hundred grams of the damp sand just as it is received
at the laboratory are weighed out and put into a jar or a graduate.
If desired, the quantity of this may be measured and a given
bulk, such as 100 c. c, may be used instead of the fixed weight.
If thus measured by volume the weight is also determined. Water
is added and the mixture is shaken violently and stirred for 2
minutes. Then the dirty water is poured off into a separate

Digitized by

Google

Report on Tests for Concrete Materials. 481

vessel. More water is added and the operation is repeated until
the water is practically clear. The water poured off is evapo-
rated, taking care not to raise the temperature much above the
boiling point, or else by another method, it is poured through
a filter paper (previously weighed) and this residue, together
with the paper, dried at a temperature of 212 deg. Fahr. The
filter paper and residue are then weighed together, the weight of
the filter paper deducted, and the remainder is considered as
silt. The percentage is recorded which the weight of this silt,
or, by the other method, the weight of the silt left from the
evaporated water, bears to the weight of the original sand.

The evaporated residue, or else the filter paper with its
residue, are ignited in a crucible at a red heat, and the loss of
weight by this process (after allowing for the weight of the filter
paper) is taken to indicate the amount of organic or vegetable
matter. The percentage of this is expressed both in terms of
the silt and of the total sand.

Blank Forms for Reports.

Forms are appended to this report that have been used in
laboratories of members of your Committee for special tests of
aggregates, see Figs. 1-4.

Respectfully submitted,

Sanford E. Thompson, Chairman,
Cloyd M. Chapman,
William B. Fuller,
Russell S. Greenman,
Arthur N. Talbot.

Digitized by

Google

482

Repobt on Tests fob Concrete Materiaia.

SANO TEST REPORT

grJ^»Ji.M

rijEAi^wjf

FIG. 1. — BAND TEST REPORT USED BY WBSTINGHOUSE, CHURCH, KERR

AND COMPANY. CLOYD M. CHAPMAN, ENGINEER IN CHARGE,

NEW YORK, N. Y.

Digitized by

Google

Report on Tests por Concrete Materials.

483

Contract No.

TESTS OF SAND fpom bank at ^ , N. Y.

Proposed for use on contract No , Ree. No Canal, Division.....

Contract sample No taken , received at Laboratory , made up

Sand is maitUy quartz and feldspar with some hornbUnd and magnetite.
Percentage of Voids, 36^; Loam. S.2; Organic matter, trace.
Parts of sand to cement, by bulk: — S sand to 1 cement.
Per cent water used -\-13.

Cement used in tests .

For test of cement see Vol. , Page
Temperature, Fah. when mixed, Air , Water

Briquettes kept in air 24 hours and then inmiersed.

TENSILE STRENGTH (in pounda per square inch)

Natural Sako

Washed Sakd

SIZE OF SAND
Paisxno Suyb

Briquette' _ .
N©. 7 days

178
186
192
192
170

Total I 918
Average, 184

28 daysl

252 I'

274 1

2S0 \\
268

260 ,|

Briauette

1294
269

Total
A verage

7 days l28 days,
182 ! 286

196
190
198
196

962
192

276
264
262
274

I No.

' 6(1/
]10

130

Per H
cent.

100.0 \
lOO.Tl
99.2 \
97.6 \
87.0 \\
68.0 \\

No.

100

140

200

I Per

I cent.

I SS.2

I 16.0

7.8
3.2
2.0
14

, Reported 19 .

I Examined and Approved

1362 1 1 Resident Engineer.

270 11 Accepted . Rejected.

Tests for strength made by
Tests of sand made by

. Recorded by

FiQ. 2. — Report Card of Sand Test New York State Testtno Labor-
ATORT, Albany, N. Y., Russell S. Greenman, Resident Engineer
IN Charge of Tests.

Digitized by

Google

484 Report on Tests for Concrete Materials.

SANFORD E. THOMPSON. FUb, 8WBS.

CoNBUiA'iNa Enoinbbr, Date, Feb. 18, 1911.

Newton Highlands, Mass. Volumetrie or Density Teet. Experimentor, W. O. L

( 1) Test Number ? V-e39

( 2) Date S/W/ll

By volume 1:2:4

( 3) Nominal mix By weight 1:1.8:S.8

( 4) Brand of cement

( 5) Weight of cement 2.000

( 6) Weight of aggregate parsing a No. 100 sieve 0.0007

(7) '^ " " coarstt' than a No. 100 sieve 11.180

( 8) " " vessel and water (before using) 1 .600

(9) " " " '' " (after using) 0.740

(10) " •* water used 0.760

(11) Total weight mixed (5) + (6) + (7) +(10) 1S.947

(12) Weight of trav+tools (aft«r mixing) 2.760

<13) " " '' + " (before mixing) 2.626

(14) Weight of mixed adhering 0.126

(16) Weight of waste+water 0.126

(16) Weight of waste 0.094

(17) Weight of free water O.OSl

(18) Net mix of set = (11) -(14) -(17) 18.791

(19) Water left on tray -^5J!^-(|y^^ior ^ ^^

(20) Net water set = (10) -(17) -(19) 0.696

(5)X(14)

(21) Net cement (5)" (5)^, (6)4- (10) ^ -^^^

(22) Net aggregate passing a No. 100 Sieve - (6) — (K\A.{ k\ XTim 0.0O67

(23) Depth of concrete in cylinder 0.44^

(24) Volume of concrete in cylinder 0.0869

(25) Net water per cu. ft. as mixed (10) h- (24) 8.760

(26) " " " " set (20) -^ (24) 8.020

(27) Net cement per cu. ft. as set (21) ^ (24) 22.000

(28) Net aggregate per cu. ft. as set (22) -f [(7) -f- (24)] 128.60

(29) Abs. vol. of water per cu. ft. as set (26) -^62.3 0.1286

(30) Abs. vol. of cement (27) ^62.3 0.11S9

(31) Abs. vol. of aggregate (28) -^^^.5 0.760

(32) Abs. vol. total* (29) + (30) + (31) / .0024

(33) Weight of form +concrete 19.000

(34) " " " 6.166

(35) " " concrete 13.844

(36) Temperature of water 70'*F-.

(37) Time of mixing 10,00 a.m.

(38) Remarks on consistency JeUy like

Pig. 3. — Mechanical Analysis Report, Sanford E. Thompson, Con-
sulting Engineer, Newton Highlands, Mass.

Digitized by

Google

Report on Tests for Concrete Materials.

485

8ANFORD E. THOMPSON, FUe, 8WBE.

CoNBULTiNa Enoxnesb, Date, Jan. £4$ 1911.

Newton Highlanda, Mam. Mechanical Analysts. Experimenter, W. E. S.

Test made for SWBE,

Description Gravel from Neponsel River.

Samples Taken (date) Jan. 23 ^ 1911, Shipped to laboratory in bag.

Samples Received (date) Jan. 24, 1911,

Weight Total Sample 20 lbs.

Client's Mark

Laboratory Mark A, S., 186 SWBE.

Analysis No A. S. 186.

Per cent. Moisture 1.6%

Sise of Sieve

Total

Totol

Per
cent.

Totol

Per
cent.

Total

Inches

No.

Weight
PasSng

Per cent.
Passing

Finer
than H"

2.50

H
1
J

6
12
20
40
50

100

200

2.00

1.60

998.0

677.2

669.0

388.0

366.0

160.6

64.2

17.0

11.8

6.8

3.6

100.0%
67.8
57.1
38.9
36.6
16.1

6.4
1.7
1.2
0.6
0.36

1.00

0.50

.... j ... .

0.25

100.0%
94.3
38.8
16.6

4.4

3.1
1.6
0.9

0.16

0.0583

0.0335

. . .

0.0148

....

0.0110

0.0055

::::i::::

0.0030

Remarks

Washing No. W226
Tensile No. T926
Report made Fe&. 1, 1911.

Approved,
Noted by S.E.T.,

Date Initials

Jan. 24, 1911. W. O. L.
Jan. 28, 1911. S. E. T.

Fig. 4. — ^Volumbtrig or Density Report.

Digitized by

Google

AGGREGATES FOR CONCRETE.
By William M. Kinney.*

Within the past few years has been seen the almost universal
adoption of the Standard Specifications for Portland Cement f of
the American Society for Testing Materials. Practically all
cement made in this country is guaranteed to meet these require-
ments and it is only occasionally that a shipment fails to pass the
specifications by a safe margin. Despite the precautions taken in
the making and testing of cement, however, there are occasionally
complete or partial failures of concrete work. Fortimately these
failures usually occur during the process of construction so that
the loss of life and property is relatively small, but that there
should be any loss whatever in this type of construction should
lead such bodies as this Association to strive for greater efficiency
by a careful and thorough study of the materials entering into
and the workmanship required for concrete.

It is seldom possible to determine positively the cause for
such failures; in fact, the number of reasons given for a particular
failure usually varies with the number of engineers employed on
' the investigation. It is essential, therefore, that in the construc-
tion of any concrete or reinforced concrete structure, that each
step be made with absolute surety in order to attain success.

Deficient strength in concrete is usually due to one or more
of the following causes: poor workmanship, unsatisfactory
aggregate and unfavorable weather conditions, assuming of course
that the design is right and that the cement had been carefully
tested to the standard specifications. Poor workmanship can be
eliminated from this discussion, because with the widespread
distribution of literature on the mixing and placing of concrete,
it should be possible for any contractor or user to handle concrete
in such a manner that success will be assured. However, the
question of aggregate and temperature conditions, with usually
a combination of the two, is a matter deserving of very careful
consideration. Of these two, the question of aggregates is the

* Assistant Inspecting Engineer, Universal Portland Cement Company. Pittsburgh, Pa.
t Standard No. 1. National Assooiation of Cement Users. — Ed.

(486)

Digitized by

Google

Kinney on Aggregates for Concrete. 487

paramount issue, as it is almost a certainty that if good, clean
aggregates were used, at least three-fourths of the frozen concrete
would be eliminated. The reason for making this statement is
based upon the fact that when concrete is subjected to actual
freezing weather during the early stages of hardening, it is usually
well protected and poor work seldom results; but when it is
deposited at a temperature above freezing it is not protected, and
owing to delay in hardening due to unsatisfactory aggregates,
freezing may occur the following night or even a number of
days later. It is certain that in a great many cases had clean
aggregates been used the concrete would have hardened suffi-
ciently during the favorable weather to have withstood the sub-
sequent freezing without resultant injury to the concrete.

The aggregates, therefore, play an important, if not the most
important part in concrete work, and are without doubt subject
to the greatest variation, yet have received up until the present
time the least study of all the adjuncts of good concrete. To be
sure, there are the studies of Feret, Candlot and other European
experts. Likewise, the Structural Material Division of the
United States Geological Survey (now under the Bureau of
Standards) and individual investigators, such as Thompson,
Spackman and Greenman have given valuable information
obtained from their study of sand, gravel and other aggregates.
This work, however, has of necessity been limited somewhat to
the study of deposits local to the various laboratories, and the
results are chiefly valuable in giving an idea of the most advan-
tageous tests for detecting the properties of a particular aggregate,
which render it good or poor for concrete. It remains to apply
these tests in the study of the aggregates being used or proposed
for use in individual structures or in particular localities.

The report of the Committee on Specifications and Methods
of Tests for Concrete Materials has been presented. It would
seem highly proper for this committee at the earliest possible
date to confer the committees on Concrete and Reinforced Con-
crete of the American Society for Testing Materials, American
Society of Civil Engineers, and Association of American Portland
Cement Manufacturers, looking toward the establishment of a
standard set of tests for concrete materials. Having outlined such
methods, it should then be within the power of this committee to

Digitized by

Google

488 Kinney on Aggregates for Concrete.

influence the various Government, State and commercial bodies
interested in this subject to inaugurate at once an extensive study
of these materials by the methods outlined.

The magnitude of such an investigation may lead to the
thought that it is not feasible, but there is no reason why, if
properly organized under an advisory board composed of rep-
resentatives of the various large engineering societies, the work
cannot be systematically done without duplication of tests by the
Bureau of Standards and the various State Universities and
Experiment Stations together with whatever assistance may be
rendered by other laboratories of a public or private nature.

In 1905 there was established at St. Louis a laboratory under
the Structural Materials Division of the United States Geological
Survey, which published in 1908 Bulletin 331 on Portland Cement
Mortars and Their Constituent Materials, Neither this laboratory
nor its successor in this line of work, the Bureau of Standards,
has given any further data on this very important subject. It
is to be hoped that the Bureau of Standards will continue this
very necessary work, as it would be unfortunate to lose the prac-
tical information gained by the investigators in conducting the
j&rst series of tests. Good work has also been done in several of
the State Universities and Experiment Stations. It is pleasing
to note that the University of Wisconsin is to start at once under
Mr. M. 0. Withey, a comprehensive study of the aggregates of
Wisconsin. If Wisconsin can do this, why cannot other States
be interested in a similar investigation?

Such an investigation will not by any means solve the prob-
lem and insure good work in the future, but it will give an idea
of the relative merits of the various aggregates available for con-
crete in any particular locality. This in itself will be a most
important step forward, as it is safe to say that less than half of
the architects and engineers in this country know the crushing
strength that can be obtained from the mixtures they specify
with the aggregates that are being used on jobs under their super-
vision every day.

This statement was recently borne out by an investigation of
the aggregates available for concrete in one of our larger cities
prior to the formation of a Building Code for concrete and rein-
forced concrete buildings. In this code it was originally planned

Digitized by

Google

Kinney on Aggregates for Concrete.

489

to have the requirement for compressive strength of 1:2:4 con-
crete at 28 days at least 2400 lb. per sq. in. This was later
reduced to 2000 lb. and the allowable stresses figured accord-
ingly. An investigation by the writer and others interested in
the writing of the Code developed the fact that though the supply
of sand and gravel was limited to practically one source and the
crushed stone to several quarries in the same vein, very few tests
on the aggregate were obtainable and these seemed to indicate
that even 2000 lb. was too high. A series of tests was started
and the results to date are quite surprising. The specifications

Table I. — Comparative Compressive Strength in lb. per sq. in. of
Concrete Made from Various Aggregates.

Mark.

7 Days.

Ordinary gravel..
Large gravel

Stone

677
o9o
603

626

898
677

874

816

819
858

962

879

28 Days.

3Mos.

1578
1390
1633

1512
1673
1943

1533

1709

1626
1640

1684
1779

1580

1649

1582

1704

1824
1972

1880
1888

2031
2206

2484
2240

Proportions.

1 part typical cement.

2 parts nver sand passing */u in. screen,
'cl Vw in. ' "' '

4 parts river gravel

. to V« in.

1 part tsrpical cement.

2 parts nver sand — 3 parts through Vi«

in. screen, 1 part Vi« in. to 1/4 in.
4 iMtrts river gravel 1/4 in to 1 in.

1 part typical cement.

2 parts nver sand — 3 parts throujgh */i«

in. screen, 1 jmrt '/is in. to 1/4 in.
4 IMtrts crushed stone 1/4 in. to 1 in.

Typical cement used was a mixture of five brands. Medium consist-
ency. Hand mixed. Test pieces 8 in. in diameter and 16 in. long. Aged
in temperature of about 70^ F. and protected from drying out by cotton bags
wet twice a day.

require that the coarse aggregate shall pass a 1-in. ring and be
retained on a i-in. ring; fine aggregate to be all that passing a
i-in. ring. The material produced commercially was being
screened through a |-in. and over a A-in. mesh, so that tests were
made on the aggregates as conmiercially produced and on samples
specially prepared to meet the specifications. In the case of the
sand, this was done by mixing with three parts of sand passing a
^-in. screen, one part of the fine material passing a J-in. screen
obtained from the i^-in. to f-in. gravel. The results up to three
months are shown in Table I and prove conclusively that with

Digitized by

Google

490

Kinney on Aggregates fob Concrete.

Tabub II.— Tests on Typical Ceiibnt used in Coiipbessiqn Tests.

Fineness 100 mesh — 94.8 per cent. Initial Set — 4 Hours.

Fineness 200 mesh— 77.4 per cent. Final Set— 7 Hours 25 Minutes.

Soundness — Satisfactory. Normal Consistency — 24 per cent.

Chemical Analysis.

Silica 20.72

Alumina 7.24

Iron Oxide 2.84

Caleium (hdde 62.85

Magnesia 2.47

Sulphuric Anhydride 1 . 42

Moisture and undetermined 2 . 46

Tensile Strength in lb. per sq. in.

NeM.

1 : 3 Ottawa Sand.

24 Houni.

7 Days.

28 Days.

3 Months.

3 Days.

7 Days. | 28 Days.

3 Months.

310
815
330

600
660
660

740
780
750

765
775

740

180
190
200

340
300
300

420
390
400

440
480
460

318

640

757

760

190

313

403

460

I :

3 Commeroial Sand.

3 parts throi

7 Days.

28 Days.

3 Months.

7 Days.

240
300
265

268

390
380
36.^

378

390
425
380

39S

290
340
290

307

1 : 3 Special Sand.

28 Days.

460
420
400

427

3 Months.

505
480
535

506

Granulombtric Analysis.

Commercial
Sand.

'^t»

P«r fwnt retained on No. 4 mcvh . . t . . . . . . t . . . . t - - , - 1 - - -

10.6
19.2
25.8
11.8

5.8
24.4

0.6

1.4

10 **

31.5

20 '•

14.5

30 •*

15 2

40 "

10.5

50 "

4.2

80 '*

18.5

100 "

2.3

Through 100 "

3.2

Digitized by

Google

Kinney on Aggregates for Concrete.

491

such proportions a requirement of 2000 lb. per sq. in. is too hi|^
for the local material, and that either the amount of cement will
have to be increased or the proportions readjusted so as to pro-
duce higher results. In these tests, which were made on 8-in.
diameter cylinders 16 in. long, the concrete was accurately pro-
portioned and mixed by hand and the test pieces were stored in
an even temperatiu-e approximating 70* F., being protected from

Table III. — Compressive Strength in lb. per sq. in. of Concrete
CTLifn)ERS Using Material from Same Source as Table I.

Brand of Cement.

10 Days.

4«6
498
600

521

439
539
409

462

620
527
515

520

395
521
536

484

430
396
434

420

30 Days.

3 Months.

803

960
828

Ill

893

1275

990
904
800

1182
1096
1018

898

1098

820

999

1142

1433
1172
1136

987

1247

774

872
898

1138
1263
1316

848

1235

706
834
1084

1263
1219
1160

878

1214

6 Months.

looa

1185
1131

1106

1152
1063
1299

1171

1333

949

1464

1248

1084
1382
1609

1358

1060
1064
1046

1063

Proportions: 1 part cement (5 brands used individually), 2} parts river
sand through A in., 5 parts river gravel A »n. to li in.

Test pieces 8 in. in diameter and 16 in. long, mixed with batch mixer,
stored in open shed and sprinkled night and morning for first 7 days.

drying out by cotton bags wet twice a day. The cement used
was a mixture of five representative brands, the results of tests
on which are shown in Table II. In the same table are granulo-
metric and tensile tests on the commercial and special sand. It
will be noted that despite the fact that the commercial sand
approximates very closely the tensile strength of Ottawa sand and
the special sand exceeds in strength that of Ottawa sand, yet

Digitized by

Google

492

Kinney on Aggregates for Concrete.

used in concrete with gravel from the same source and a very
good grade of crushed stone, neither of these materials developed
a crushing strength of 2000 lb. per sq. in. at 28 days. Other

Table IV. — Comparative Cold Weather Tests on (8 in. Diameter by

16 IN. Long) Cylinders Mixed 1 Cement, 2i Sand, 5 Gravel,

December 4, 1911. Cyunders Protected from Frost by

Cotton Bags. Hand Mixed, Medium Consistency.

Crushiiig Strength in lb. per sq. in.

Age.

3 days in air and 1 day in laboratory .

7 days in air.

14 days in air.

30 days in air.

Brand "A"

116
122

119 Average

227
265

246

4S4
458

471

640
739

689

Brand "B"

95
95

95 Averago

199
209

205

440

388

414

495
653

574

Temperature — Degrees Fahrenheit.

Ist day

2d "

3d "

4th "

5th '•

6th "

7th ••

8th *•

9th ••

10th '•

11th "

12th "

13th "

14th "

15th "

Max.

30
36
48
54
54
54
66
61
59
51
42
47
58
38
33

Min.

18
21
27
29
40
48
46
58
45
35
32
39
38
32
30

Mean.

24
28
38
42
47
51
56
60
52
43
37
43
48
35
32

16th day.
17th ■
18th
19th
20th

2l8t

22d
23d
24th
25th
26th
27th
28th
29th
30th

Max.

Min.

Mean.

results obtained on material from the same source are shown
Tables III and IV, and the latter table particularly shows^that
this aggregate is poor for cold weather work. Great care'should
be exercised in the use of such material in the winter time as

Digitized by

Google

Kinney on Aggregates for Concrete. 493

unless the materials are heated and the concrete well protected,
poor results are almost inevitable.

Having established the value of aggregates obtained from
any particular soiu-ce, it is then necessary to see that the mate-
rials received on the job are equal in quality to the samples
tested. An examination of the bank, pit or quarry may often
reveal the fact that it would be impossible to obtain a uniform
product. In such cases, extreme care must be exercised by the
men obtaining the material and frequent tests should be made
to see that that received is right.

The pier shown in Fig. 1 is a good illustration of the effect of
variation in aggregate. This pier was built in connection with
two abutments to support a steel girder railroad bridge across a
small stream. Prior to starting the concrete work the engineer
secured a sample of the sand and gravel which the contractor
proposed to use, and the laboratory results obtained indicated
that the materials as sampled were entirely satisfactory for con-
crete work. That the material as received was not good, and
apparently not equal to the sample, is evidenced by examination
of the condition of the concrete. A photograph of the gravel
bank (Fig. 2) reveals the seat of the trouble. The dark streaks
are apparently decayed vegetable matter, while here and there
through the bank will be noticed strata of very fine uniform size
sand. Examination of such a bank indicates quite conclusively
that the aggregate could not rim uniformly, and imdoubtedly the
delay in hardening which finally resulted in freezing was due to
the presence of too much fine sand and loam. That this was
the cause of the trouble is conclusively proven by the fact that
the poorest results are evidenced at the water level where the
fine material held the moisture and at the end of a day's work
about midway of the pier where most of the fine material floated
to the top of the concrete.

On another job of a similar nature, but involving the use of
approximately 50,000 barrels of cement, the results were anal-
ogous. A large amount of concrete was condemned on account
of imsatisfactory strength. In the middle of a 1000-yd. founda-
tion laitance was found over a foot thick which had the appear-
ance of wet clay and could be readily dug out with a knife even
though the concrete had been in a month. Run-of-pit gravel was

Digitized by

Google

494

Kinney on Aggregates for Concrete.

being used which had been tested for percentage of fine and
coarse material and found satisfactory, but examination of the
pit revealed the fact that it would have been impossible to have
secured a sample which would have fairly represented the deposit.

Fia. 1. — DISINTEGRATED CONCRETE PIER, DUE TO POOR AGGREGATES.

In some places there was very little sand, while in others there
was practically no gravel. As a mixture of one part cement and
seven parts run-of-pit gravel was being used, the natural result
of the use of the latter material can be imagined. Further than

Digitized by

Google

KiNNET ON AgGRBGATES FOR CONCRETE.

495

this, there were strata of very fine sandy loam throughout the pit
and the large section of laitance was the result.

The foregoing are a few of the many examples which can be
cited where failure to appreciate the value of .having good
aggregate has led to trouble and such examples emphasize the
importance of tests of the aggregates. It is quite surprising to
find architects and engineers basing their designs on a certain
strength of concrete which cannot be obtained or may be largely

FIG. 2. — GRAVEL BANK FROM WHICH POOR AGGREGATE WAS OBTAINS^.

exceeded in actual practice, and surely our work is devoid of the
fundamentals of good engineering when we use without ^scrim-
ination aggregates of high and low strength giving values. A
more thorough study of our sources of supply for concreting
materials and a more careful inspection of these materials as
they are received on the work, is of utmost importance to the in-
dustry and it should be one of the first efforts of this Association
to study this important subject.

A partial bibliography of the literature on aggregates for con-

Digitized by

Google

496 Kinney on Aggregates for Concrete.

Crete is given below with the hope of furthering study on the
subject.

Concrete Aggregates, by Sanford E. Thompson, Proceedings National
Association of Cement Users, Volume II, p. 27. See also Engineering Record,
January 27, 1906.

Sands and Their Relation to Mortar and Concrete, by Henry S. Spackman
and Robert W. Lesley. Proceedings, American Society for Testing Materials,
Volume VIII, p. 429.

The Value of Sand in Concrete Construction, by E. S. Larned, Proceedings,
National Association of Cement Users, Volume IV, 205.

Concrete — Its Constituent Materials, by Russell S. Greenman, Barge
Canal Bulletin (New York State), November, 1909, p. 429.

Practical Tests of Sand and Gravel Proposed for Use in Concrete, by Russell
S. Greenman, Proceedings American Society for Testing Materials, Vol-
ume XI, p. 515. See also Engineering Record, Volume LXIV, No. 3, p. 66.

Economical Selection and Proportion of Aggregates for Portland Cement
Concrete, by Albert A. Moyer, Engineering-Contracting, Volume XXXIII,
No. 3, p. 52.

Concrete Aggregates by Dr. J. S. Owens, Concrete and Corutrudional
Engineering^ March 1909.

\Portiand Cement Mortars and their Constituent Materials, by Richard L.
Hui|S)plhrey and William Jordan, Jr., BuUetin 331, United States Geological
Survey.

Good Concrete and How to Get It, by F. M. Okey, Municipal Engineering,
May, 1909.

• Notes on Concrete, a discussion printed in Journal of Association of Engi-
neering Societies. See Engineering Record, Volume LXI, No. 5, p. 125.

Impurities in Sand for Concrete, a discussion, — ^Transactions American
Society of Civil Engineers, September, 1909.

BUist Furnace Slag in Concrete, booklet published by Carnegie Steel Com-
pany, Pittsburgh, Pa.

Tests t0 Determine Effect of Mica on Strength of Concrete, by W. N. Willis,
Engineering News, February 6, 1908.

Directions and Suggestions for the Inspection of Concrete Materials, by
Jerome Cochran, Engineering-Contracting, Volume XXXVII, No. 5, p. 116.

A Study of Sand for Use in Cement Mortar and Concrete, by E. S. Lamed,
Journal Association of Engineering Societies, Volume XLVIII, No. 4, April,
1912.

Digitized by

Google

DISCUSSION.

President Humphrey. — The relation of the sand to the Mr. Humphrey,
mortar, especially of the strength of mortars made of the com-
mercial sand to that of mortars of standard Ottawa sand, is of
interest. There has been a great deal of debate in the Joint
Committee as to whether they should show the same strength
or possibly less. Engineers from New England, especially from
Boston, object to requiring the same strength, because in their
locality it is impossible to obtain a sand that would show this
strength, therefore making the requirement a hardship. I believe,
however, with Mr. Kinney, taking sand the country over, that the
various sands should show at least the strength of the Ottawa
sand, and that localities where it is not possible to obtain this
strength must fix the requirement to suit their locality. It is
certainly a fact that the well graded sands give higher results
than the standard Ottawa sand. The standard Ottawa is a one
size sand with a large percentage of voids and the strength of
the mortar in which it is used is less than that of sands well
graded.

Mr. Kinney states that grading gives density and that the
increased density gives an increased strength. The retaining
walls with the disintegrating mortar referred to are as fine an
example of cause and effect as could possibly be had. It did not
require a technical man to look at the sand bank to see that the
material was not suitable for mortar. It was quite evident that
the fine sand in the mortar retarded the hardening, and left the
mortar with insufficient strength and thus easily damaged by
frost action. This is one of the most important subjects that we
can discuss. There is more information desired and great need
for intelligent understanding of just what important part sand
plays in mortars and concrete; it probably has a more important
bearing on the resulting strength of the structure than the strength
of the cement. I think we are more prone to ascribe the defects

(497)

Digitized by

Google

498

Discussion on Aggregates.

Mr. Kinney.

Mr. Wilson.

Mr. Httmphrey.

in mortar or concrete directly to the cement rather than the sand
used.

Mr. Wm. H. Kinney. — The sand from this particular bank
had been tested by the engineer and was found to have a strength
which compared favorably with Ottawa sand. In fact, it passed
the 70 per cent strength requirement by a very satisfactory
margin. However, the samples did not fairly represent the
material the contractor was obtaining from the bank.

Mr. Percy H. Wilson. — There was a case called to my
attention 5 or 6 months ago where a railroad engineer endeavored
to obtain in the laboratory a compressive strength of 2,000 lb.
per sq. in., the requirement of the Joint Committee. On obtaining
only about 1,800 lb. the test was tried over and over, varying
the aggregate, but the strength never w^ent up to 2,000 lb. Now
is this requirement of the Joint Committee too high?

President Humphrey. — Many contractors and probably
a good many engineers and architects never seem to think it is
necessary to test the sand. They simply pass on it from a visual
inspection, whereas the committee states that a differentiation
between a good or a poor sand can only be made by actual test.

I would say I do not think the standard requirement set by
the Joint Committee of 2,000-lb. concrete is too high. It is a
standard we should work to. Unfortunately there is the practice
in this country of specifying the proportions such as 1 : 2 : 4 or
1:3:6. Such proportions mean absolutely nothing, as the
proper portions of materials cannot be determined until the
qualities of the material itself are known. The voids vary with
all the material all over the United States. A study of Bulletin
No. 331* on the subject of sand will show that there is a wide
range of variation, first, as to the size of the particles, and second,
as to the hardness and character of the material itself. A traprock
with a high compressive strength used as an aggregate in a certain
proportion will give a concrete of much higher strength than an
aggregate of very soft limestone. If a contrac^tor with a fixed
proportion cannot obtain 2,000 lb. or with any variation of the
proportions, then more cement must be used or another aggregate
obtained. The mere fa(!t that his material in standard propor-
tions will not give 2,000 lb. in my mind does not mean that he

♦U. S, Geological Survey.

Digitized by

Google

Discussion on Aggregates. 499

should not try to get 2,000 lb. There are too many structures Mr. Humphrey,
being erected in this country assumed to have a strength of 2,000
lb. and upwards where as a matter of fact the strength is very
materially less. The sooner more attention is given to the strength
of the concrete as well as of the cement and the steel, the better
the results will be.

The Joint Committee report specifies that the relation of
the cement to the aggregates shall be 1 to 6. The next report
will contain a table showing the strength of concrete from aggre-
gates of different character and proportions, so that for any
aggregate, say a soft limestone, reference to the table will show
approximately the proportion of cement and aggregate necessary
to obtain a 2,000-lb. concrete.

Mr. L. R. Ash. — In connection with the sand and grading of Mr. Ash.
aggregates suggested by the disintegrated concrete in the pier,
the water put into a mixture sometimes makes considerable
difference and also the depositing of concrete in or through water.
I want to call attention to a very interesting occurrence here in
Kansas City a couple of months ago. We have a bridge over
the Kaw River built some 4 or 5 years ago and lately during some
very intensely cold weather, the cylinder pier split open and went
down so badly that false work had to be placed to hold the bridge
up. In the search for the cause the first suggestion was that the
contractor had not put in any cement but some line that seemed
to be accessible in the neighborhood. I was talking to the fore-
man of the job and he told me the circumstances, which it seems
to me would explain it very clearly. The concrete was deposited
through several feet of water with the result that there were layers,
at intervals, of material more like laitance than anything else,
which was chalky in its consistency and would absorb great quan-
tities of water. To me this very thoroughly explained the break-
ing down of the concrete. I have frequently noticed in depositing
concrete that small pockets or irregularities in the distribution
of the water would crdate conditions that would show up badly,
and entirely outside of any irregularities in the aggregate at all.
Sometimes I think that is a matter which is overlooked in the
handling of concrete and depositing of the same in the forms.

Mr. Kinney. — I would like to ask whether anyone has noticed Mr. Kinney,
in the fracture of gravel concrete that the pebbles pull out rather

Digitized by

Google

500 Discussion on Aggregates.

Mr. Kinney. than break? This was particularly noticeable even at six months
in these building code tests and it occurred to me that it
might be a characteristic of all gravel concrete in the early stages
of hardening.

If anyone contemplates the making of test cylinders on a
job or in the laboratory, I might add that we have found a very
satisfactory form made of galvanized iron. This form consists of
a flat piece of galvanized iron 16 in. wide and long enough to come
together and be soldered to form an 8-in. cylinder. The bottom
is circular and slightly larger in diameter than 8 in. and is soldered
to the cylinder. With the use of a sharp instrument these soldered
joints can be readily opened and the forms stripped. Such forms
cost 22 cts. apiece and they are worth the price.

Mr. Humphrey. PRESIDENT HUMPHREY. — Whether or uot tcsts of Cylinders

would show a fracture through the gravel itself, depends largely
on the age and the proportions. A proportion of 1 : 3 : 6 probably
up to a year might not show such a failure. In many tests made
in St. Louis the gravel did fail, especially 1:2:4 mixtures. The
fracture was right through the particle, even the hardest particles.
The governing factor seems to be the manner in which the materials
are handled; that is, the character of the concrete and its age.
A well made gravel concrete of 1 : 2 : 4, or in which there is one
part of cement to six parts of gravel, at the end of six months
should certainly show a fracture through the gravel.

Digitized by

Google

FIELD INSPECTION AND TESTING OF CONCRETE.
By G. H. Bayles.*

In the spring of 1910 we were commissioned to design and
superintend the construction of a warehouse 100 x 210 ft., 4
stories high, in the Borough of Brooklyn, New York, for the New
York Dock Company. The use of reinforced concrete was
recommended and adopted, but from experience and knowledge
of the many criticisms of this material and the considerable num-
ber of failures in its use, it was decided that very careful inspec-
tion should be maintauied. Consequently when the work started
in the latter part of July the engineer who designed the building
was put in charge, assisted by two inspectors, both experts in the
making and placing of reinforced concrete. The engineer assumed
general charge of the work. One inspector superintended the
proportioning and mixing of the concrete and the other the
placing in the forms. They collaborated in the inspection of con-
structing and wrecking forms, placing reinforcement and testing
materials.

The results of the tests on the above work were on the whole
so satisfactory and the information obtained was deemed of such
importance that they were continued on some other work during
the summer and fall of 1911. This latter work consisted of the
reconstruction of a block of six warehouses. The whole block is
210 X 374 ft. in plan and 4 stories high. The buildings were of
the old type, brick walls with wood interior. All the wood was
removed and replaced by reinforced concrete, mak'mg the build-
ing fireproof throughout. As this work was done with greater
despatch than the first job it was thought best to have more
inspectors. Consequently a chief inspector was put in charge of
the whole work, assisted by one inspector on the proportioning
and mixing, one on the placing concrete and one solely for making
tests. This arrangement worked very satisfactorily and allowed
the engineer to devote more of his time to other work.

There seemed to be no recognized system of testing concrete

* AraiBtant to J. W. Galbreath. ConBulting Engineer, New York. N. Y.

(501)

Digitized by

Google

502 Bayles on Inspection and Testing of Concrete.

or concrete materials on the job, so a series of tests was arranged
to meet what were believed to be the necessary requirements.
These tests were for cement: fineness, constancy of volume, time
of setting, initial and final, specific gravity and crushing strength
of 1: 2 cement mortar; for sand: tests for loam, fineness and
percentage of voids; for stone: tests for percentage of voids;
for concrete: tests of the crushing strength. No tests of the
reinforcement were made on the job.

A temporary laboratory was built near the work and the
necessary apparatus installed. This laboratory was intended to
Ix* such as could be built on any job and no arrangements were
made for heating it except the use of a small oil stove in extreme
weather. It was 12 x 8 ft., built of |-in. sheeting on 2 x 4 in.
frame ^vith felt roof; one door front, and windows one side and
rear.

cement tests.

At first 6 samples for testing were taken from every car of
cement, but later this number was reduced to 3. For fineness
the usual 100-mesh and 200-mesh sieves were used. The inspector
shook the sieves in his hands and judged by the eye when no
appreciable quantity of cement was going through. Scales
graduated to milligrams were used to determine the proportions.
A nearby tinsmith made the boiler and wire rack for the con-
stancy of volume tests and the glass plates were cut on the job
from scrap window glass. The boiler is 15 in. high and 8 in. in
diameter with a lid, and the rack was made to hold 12 pats at a
time. The test pats were boiled for 5 hours or more. The Vicat
needle was used to determine the time of setting. The variable
temperature made the remits of these tests vary so greatly as to
be of little value. Le Chatelier's specific gravity apparatus was
used with gasoUne for determining the specific gravity.

The test for crushing strength, not being in such general use
as the other tests, requires more detailed description. It consisted
in testing to failure 4-in. cubes of 1 : 2 cement and sand mortar.
More elaborate apparatus was necessary for making these tests.
For this purpose were required gang molds for the cubes, a damp
closet large enough to hold 4 gang molds, pans of water in
which to submerge the test pieces, and a compression testing

Digitized by

Google

Bayles on Inspection and Testing of Concrete. 503

machine. The gang molds, Fig. 1, were 'made of 1-in. lumber
held together by iron clamps. The measm-ements were not exact,
but the surface areas were approximately correct within the limits
of accuracy of the tests. While only 2 cubes were used for each
test the gang molds were made for 4 cubes, as the same molds
were to be used for the concrete tests. They were made by a
carpenter on the job and the clamps were made by the com-
pany's blacksmith.

The damp closet was built into the comer of the laboratory,
the sides and floor of the laboratory forming two sides and the
bottom of the closet; the other sides and top were of J-in.
tongued and grooved sheeting. Two shelves were put in the

1 r

■' •-

■^ '"

^ 4' "

-«- 4^

L.4

§6

^.--^

Epj

1 1

FIG. 1. GANG MOLD FOR 4-IN. CUBES.

closet, each large enough to hold 2 gang molds, and space was
left below for a large pan of water. When a mold with test
pieces was placed in the closet a wet coffee sack was thrown
over it, the ends of the sack dipping into the water below. A
better arrangement would be to have the closet lined with felt,
the lower edge of the felt dipping into the water. The simpler
method was used as being one easily provided on any job.

The pans for water in which the test pieces were to be
submerged were made 28 x 28 in. and 5 in. deep, large enough to
hold 36 cubes each. The same size pan was used in the damp
closet. Five pans were required and they were made by one of
the company's sheet metal workers.

Digitized by

Google

504 Bayles on Inspection and Testing of Concrete.

The selection of a compression testing machine was more
difficult. Most machines designed to do this kind of work are elab-
orate and expensive. A simple and comparatively inexpensive
machine was desired and after careful investigation a hydraulic
machine was acquired, consisting essentially of two horizontal
plane surfaces between which the test pieces are crushed. The
top surface is fixed, being held rigidly in place by two side posts
extending upward from the base. The lower or movable surface
is fixed rigidly on top of a 5-in. cylindrical ram with cup leather
packing moving in a copperlined cylinder. The pressure is
applied by means of a hand pump and is measured on a gauge
reading total tons on the 5-in. ram. The gauge is graduated from
to 60 t. and while the rated capacity of the press is 50 t.,
cubes were crushed which required as much as 60 t. pressure.
As many of the newer cubes crushed at low pressure and «xact
readings on the gauge that was furnished on the machine were
difficult, a second gauge was attached reading up to 10 t. only.
When cubes requiring a greater pressure were being tested this
second gauge was shut off by a valve.

On account of the high values which were obtained at the
beginning of the work suspicion was directed towards the accuracy
of the machine, but comparisons of the cubes by testing on an
accurately gauged machine at the testing laboratory of Columbia
University showed that the small press used on the work was
correct within reasonable limits* (see Table IV).

For each test 2 cubes of 1 : 2 cement and sand mortar were
made. The sand used was taken from that used on the work
and was washed clean. To insure uniformity a sufficient quan-
tity for all the tests was taken at one time and stored in a bin
until required. For each test a little in excess of 128 cu. in. of
sand and half as much by volume of cement were used. The
sand and cement were mixed dry, then sufficient water added to
make a fairly wet mixture, that is, one that was readily puddled
but would not pour. The whole was then well mixed and placed
in the molds and puddled ^vith a small trowel. All the operations
were done by hand. As soon as the mortar was sufficiently set
to hold the markings the date and hour were marked on both

* A dcBcriptiou of a test of a similar machine is given in Engineering Neva, February 11,
1909, p. 167.

Digitized by

Google

Bayles on Inspection and Testing op Concrete. 505

cubes and the mold placed in the damp closet. At the end of
24 hours the mold was taken from the damp closet, the forms
removed and one cube crushed. The other cube was submerged
in a pan of water where it remained for 6 days and was then
crushed. All were tested to failure.

To prevent little inequalities of the surfaces of the cubes
from seriously affecting the results three thicknesses of blotting
paper were used top and bottom of each cube in the press. After
some time it was discovered that the long sides of the gang molds
were more nearly parallel than the others and thereafter care was
taken to use the sides of the cubes formed by them as the crush-
ing faces. By this means better and more nearly uniform results
were obtained.

SAND tests.

Every scow load of sand was tested for percentage of loam.
If the sand in different parts of the scow appeared to be of dif-
ferent quality more than one test was made. This test consisted
in taking a quantity of sand and drying it thoroughly on a stove.
A thousand grams of it was then taken in a 12-quart pail and
washed by turning on a hose, giving the water as it flowed from
the hose just sufficient velocity to keep the mass stirred up and
moving, the loam being carried off in the overflow. When the
water ran clear the sand was again dried and weighed, the dif-
ference in weight giving the percentage of loam. Sand contain-
ing more than 2 per cent was rejected.

Tests were also made to determine the comparative fineness
of the sand. These were only made when there appeared to the
eye to be a difference in the grade. The test consisted in taking
a measured quantity of sand (by weight)' and screening it by
hand through 20-mesh and 30-mesh sieves. These tests showed
on an average about 21 per cent retained on the 20-mesh sieve,
27 per cent on the 30-mesh sieve and the remainder passing
through the 30.

Occasional tests for percentage of voids in the sand were
made as follows: A comparatively large amount of sand was
measured and weighed to show the average weight per cu. cm.
Then a quantity by weight corresponding to 50 cu. cm. was put
into a glass graduate and the volume of water displaced meas-

Digitized by

Google

506 Batles on Inspection and Testing of Concrete.

ured. The average volume of voids was thus found to be about
40 per cent.

stone tests.

As the grade of the crushed stone varied but slightly only
occasional tests were made to determine the percentage of voids
in the stone. To do this a quantity of stone was submerged in
water for 2 hours to permit it to become thoroughly saturated.
The water was then poured off and the stone exposed to the air
half an hour to permit the surface moisture to evaporate. A
vessel of known capacity (nearly 10,000 cu. cm.) was then filled
with the stone and weighed. Sufficient water was added to fill all
the voids and the weight taken again. The difference in weight
showed the percentage of voids. As there was excess of voids
around the sides of the vessel this test is only comparative and is
not considered particularly important as the inspector can best
judge by the appearance of the batch when the voids are prop-
erly filled. For the stone used on this work the voids measured
by the above method averaged nearly 50 per cent. Measure-
ments of concrete in place and stone on the scows indicated that
the volume of stone used was about 97.5 per cent that of the con-
crete produced.

concrete tests.

What seemed to be the most important test and the test
of most practical value was that of the crushing strength of the
concrete itself as it was placed in the forms for the building.
For these tests the same apparatus was used as for the mortar
tests. The preparation of the test pieces was simpler. The
gang mold for four 4-in. cubes was taken onto the work where
concrete was being poured. The concrete was taken from the
buggies as it was being poured into the forms and was imme-
diately placed in the cube mold and puddled with a small trowel.
There was no excessive puddling as it was intended to approx-
imate working conditions as nearly as practicable. One and
sometimes two tests were made every day concrete was poured.
From this time on the operation was similar to that for testing
mortar. After 24 hours in the damp closet the forms were
removed, 1 cube crushed and the 3 others submerged in a pan of

Digitized by

Google

Bayles on Inspection and Testing of Concrete. 507

water. Of these 3 one was crushed at the end of 7 days, 1 day
in the damp closet and 6 in water, one at the end of 28 days and
the other was removed from the water at the end of 28 days and
laid aside to be tested later.

As has already been said, no attempt was made to preserve
an even temperature in the laboratory, the object being to main-
tain as nearly as practicable the condition of the concrete in the
work, so that it sometimes happened during the winter that ice
formed on the water in the pans where the test cubes were sub-
merged. The effect of this is clearly shown in the results of the
tests. A practical use of the tests of concrete was to indicate to
the inspector as well as the contractor the safe time for wrecking
forms. This time was fixed at first at 7 days, but was afterwards
shortened to 4 days during the summer and well into the fall,
when the decreasing strength of the cubes tested caused the time
to be again extended to 7 days.

results of tests.

There is nothing notable in the results of the tests for fine-
ness, constancy of volume and specific gravity, except that they
show what may be expected from the ordinary insi>ector who is
not specially trained in making such tests and where the tests
are made under the conditions prevailing on the ordinary job.
The results of the tests for initial and final set were so varied,
due, no doubt, to the changing temperature, that they lack
value as determining the quality of the cement.

Tests of the cubes of 1: 2 cement and sand mortar were
intended to show a definite relation of strength of the cement
being tested to what could be expected from the concrete made
from the cement. Comparisons of the results of these tests with
the results of the tests of concrete were so simple and proved so
satisfactory that it is believed they show with sufficient accuracy
the practical rate of setting of the cement and that the tests for
initial and final set could be discontinued for field laboratories.

Strength of Mortar. — The 173 tests of 1 : 2 cement and sand
mortar on the first job showed an average compressive strength
of 448 lb. per sq. in. in 24 hours and 2110 lb. per sq. in. in 7 days.
On the second job, where a different brand of cement was used,
198 tests were made and the average compressive strength was

Digitized by

Google

508 Bayles on Inspection and Testing of Concrete.

585 lb. per sq. in. in 24 hours and 2100 lb. per sq. in. in 7 days.
Table I* shows in detail the results of the tests on the former and
Table II those on the latter job. The results are recorded as
found with the inspector's notes, the dates alone being omitted
for lack of space. In Table I the first 50 tests were made
between August 17 and October 3, 1910, the next 50 between
October 3 and November 21, 1910, the next 50 between Novem-
ber 21, 1910, and January 25, 1911. The remaining 23 were
made between January 25 and March 16, 1911. In Table II
the first 50 tests were made between July 11 and August 24,
the next 49 between August 24 and September 22, the next 50
between September 22 and October 21 and the remaining 49
between October 21 and November 25, all in 1911.

Strength of Concrete, — The 235 tests on concrete cubes on
the first job showed an average compressive strength at 28 days
of 3450 lb. per sq. in. and on the second job, 96 tests, 2321 lb.
per sq. in. at 28 days. The tests on concrete cubes were carried
on from August, 1910, to December, 1911, tests being made nearly
every day except during the month of February, 1911, when no
work was done on account of inclement weather, and from the
completion of the first job in May, 1911, until the beginning of
the second job in July following. During the latter part of
December, 1910, and throughout January, 1911, the sand and
water used in making the concrete were heated so that the batch
had a temperature of 60 to 65 deg. when placed in the forms.
Table III shows the complete record of the tests on concrete
cubes on the first job, and Table IV those on the second job,
both giving the inspector's notes of variations from the rule.
While there are considerable variations of strength from the
maximum to the minimum most of the cubes crushed near the
average and reference to the final results indicates that the varia-
tions were due more to the method of testing than to the quality
of the concrete, for while a cube crushed after 24 hours may have
been below the average, the cube of the same set crushed after
28 days was about as often above the average as below it and vice
versa. It seems evident that the actual strength of the concrete

* The insiK'ctor's notes give all the strength testa in tons per 4-in. cube, but (or convenience
of compaiison with other testa the results have been converted into lb. per sq. in. in printing
the tables. — Ed.

Digitized by

Google

Bayles on Inspection and Testing of Concrete. 509

Table I. — ^Tests of Cement and 1 : 2 Mortar.

Car

Number.

1
2
3

4
5

f 1
2
3
4
6
6

f 1
2
3
4

, 5

1
2
3
4
6
6

1
2
3
4
5
6

1
2
3

4
5

1
2
3
4
5
6

1
2
3

f 1
2
3
4
5
6

1
2
3

4
5
6

Percentage of Cement.

Retained on Sieve.

100 Mesh. 200 Mesh

11.250 23.800

7.150
5.870
7.210

7.050

8.260
8.300
7.455
7.400
7.080
8.020

6.980
6.130
6.300
6.070
7.170

6.830
7.050
6.790
5.700
5.450
6.960

5.950
6.600
6.270
7.110
6.500
7.000

8.370
8.300
8.100
5.720
7.200
6.450

8.400
7.450
8.100
7.900

20.075
25.010
21.050

Through
200Meflh.

64.860

72.100
68.430
71.160

20.300 71.850

20.630
21.400
18.560
23.310
21.490
22.520

22.140
20.700
21.520
23.140
19.000

20.100
25.020
22.940
22.750
21.800
22.850

20.720
21.120
21.270
24.250
22.400
17.020

20.120
19.600
19.550
17.750
19.030
18.620

19.750
19.800
19.870
20.880

70.620
69.750
73.730
71.250
70.000
68.770

70.490
72.850
72.000
70.300
73.100

73.050
67.830
69.050
71.330
72.490
70.110

i 73.250
' 72.050

72.250
' 68.550

71.100
I 75.700

t 71.470
71.800
72.000
76.380
73.600
74.800

71.600
72.650
72.000
71.000

6.950
6.490

18.900
18.300

73.900
74.700

Setting.

Initial.
Hre.

1'^

4
4

4H
4

3>5

3*4

4H
4>^

4
3

2H
'3'"

6.500
6.420
5.900
6.070
5.550
5.320

4.650

20.330
18.850
17.500
18.050
17.700
17.170

72.970 '

74.700 I

76.500

75.800

76.650

77.460

Final.
Hrs.

16.550 78.720

6.490
"5 .'850

19.050 74.490 |
18.520 '75.430

2%
3 ■

314

5H
5

7
7
3M

6>i

Crushing
Strength
Specific I of 4-in. cubes.
Gravity in lb. persq. in.

3.1

3.15
3.13
3.11
3.11
3.15

3.07
3.12
3.12
3.13
3.10
3.12

3.15
3.14
3.12
3.12
3.11

3.15
3.15
3.16
3.17
3.13
3.12

3.12
3.13
3.12
3.14
3.12
3.13

3.12

6H 3.09
'3.10'

5>i
■5"*

5'i

0>4

3.14
3.13
3.14
3.12
3.13
3.13

3.09
3.13
3.14

3.09
3.11
3.12
3.12
3.06

Boiling
5 hours.

24 hrs. 7 days.

3.09
3.12

3.10
3.11
3.11

500
563
500

1053
1250
1000
1188
1125

750
1250
657
500
500
845

875
813
813
375
500

750
750
750
1025
600
813

500
475
625
500
500
525

750
875
875
875
813
875

688
750
625
688
750
625

625
875
625

375
438
438
438
563
313

1125
1250
938
438
375
438

2125
2315

2583
3190
2350
3190
3220

1870
2500
2290
2150
2350
2230

2780
2370
2275
1940
2750

2650
2400
1975
3350
2230
2750

2625
2625
2775
2500
2310
2190

2125
2500
2500
2500
2500
2625

2.500
2750
2625
2250
2250
2370

3000
3500
2625

2065
2813
2875
2250
3130
1880

2870
2440
2500
2630
2500
3190

O.K.

O.K.
O. K.
O. K.
O. K.
O.K.

O.K.
O. K.
O.K.
O. K.
O. K.
O.K.

O.K.
O.K.
O. K.
O.K.
O.K.

O. K.
O. K.
O. K.
O.K.
O. K.
O.K.

O.K.
O. K.
O. K.
O. K.
O. K.
O. K.

O.K.
O. K.
O. K.
O. K.
O.K.
O.K.

O.K.
O. K.
O. K.
O. K.
O. K.
O.K.

O.K.
O. K.
O. K.

O. K.

O.K.
I O. K.
i O. K.

Digitized by

Google

510 Bayles on Inspection and Testing of Concrete.

Table I. — Tests of Cement and 1 : 2 Mortar. (Continued )

Car

Number.

Setting.

Initial. Final.
Hr». 1 Hrs.

Specific
Gravity

Retained
lOOMesh.

on Sieve.
200Me8h.

Through
200Me«h.

f 1
2
3
4
6

[ 1
2
3

4
6

f 1

2
3

4
6
6

1
2
3
4
6
A

8.270

29.520

72.120

3.14

8.200

19.220

72.200

3.12

7.960

18.710

'73.300

3.12

8.110
8.000

19.040
18.660

72.760
73.180

3.12
3.12

7.000

19.620

73.260 Z}i

7.600

19.010

73.260

3.08

7.680

18.620

73.670

4>i

3.13

7.499

18.400

73.880 ; 4H

3.12

7.090

18.960

73.900 ZH

55i

3.13

7.200

19.570

73.160 i 4

h^i

3.12

7.330

20.300

72.190

3.13

f 1

P 2

7.610
7.730
7.860

6.400
7.000
7.720

7.600
7.100
7.260

6.060
4.770
3.860

4.480
6.910
6.710

6.940
7.070
7.320

7.300
7.610
7.560

6.200
6.350
6.530

6.490
6.440
7.030

6.330
6.430
6.920

20.620
16.370
20.410

20.900
21.210
18.050

20.680
16.760
20.760

20.300
20.400
14.060

17.360
18.780
23.730

20.030
19.300
19.500

17.700
18.860
21.650

18 500
18.000
20.940

19.040
19.530
18.910

18.660
18.780
18.790

71.560
76.700

2H '

3.14
3.11
3.13

3.13
3.12
3.09

3.08
3.07
3.09

3.15
3.12
3.12

3.12
3.12
3.12

3.11
3.11
3.13

3.12
3.13
3.13

3.11
3.13
3.14

3.12
3.13
3.14

3.14
3.12
3.13

71.690

f 1

« ll

R 2

f 1

8 2

f 1
T 2

72.620
71.780
73.650

71.620
76.030
71.860

73.440
74.700
82.030

78.160
76.260

4H
6

I'A

6
6

6
6
6K

6«

r 1

U 2
3

f 1

V 2
3

f 1

W 2

f 1

X 2

f 1

Y 2
3

69.600

72.560
73.400
73.170

75.000
73.500
70.550

75 . 150
75.470
72.510

74.320
74.020
73.850

74.970
74.740
74.200

hV^
6K
5H

5Ji

6^4

.•■>}4

8
8

7
7
9

8
8

Crushing

Strength

of 4-in. cubes,

inlb. persq. in.

24 hrs. I 7 days.

1188
1188
1125
400
438
476

876
1126
1125
1125
1250

938

538
563
563
563
626
438

375
500
438
438
438
600

438
438
438

600
513
676

625
638
600

188
188
250

250
188
188

250
250
250

188
188
188

313
313
250

313
250

288

275
275
313

2260
2500
2630
2000
2500
2625

2625
2376
2376
2625
3250
2500

2250
2375
2600
2875
2500
3125

2437
2750
2500
2500
2875
2626

2000
1876
2065

2376
2250
2275

2500
2625
2188

2260
2250
2260

2688
2250
2313

2188
2260
2250

1875
1750
1875

2000
1937
1750

1937
1750
2375

2125
2312
2500

Boiling
5 hours.

O.K.
O.K.
O.K.
O.K.
O.K.
O.K.

O.K.
O.K.
O.K.

O.K.
O. K.
O.K.

O.K.
O.K.
O.K.

O.K.
O. K.
O. K.

O. K.
O. K.
O. K.

O. K.
O.K.
O.K.

O.K.
O.K.
O.K.

Digitized by

Google

Bayles on Inspection and Testing of Concrete. 511

Table I.-

—Tests of Cement and 1

: 2 Mortar.

{Continued )

Car
Number.

Perce
Retained
lOOMeahy

Qtage of Cement

Sett

Initial.
Hrs.

ing.

Final.
Hrs.

Specific
Gravity

Crushing

Strength

of 4-in. ctibea,

inlb. pcrsq.in.

Bofling
5 hours.

on Sieve.
200Mesh.

18.450
19.000
19.280

Through
200Meah.

74.900
73.960
73.330

24 hrs.

313

438
250

7 days.

f 1

Z 2

6.630
7.030
7.240

5>i

8H
6>i

3.13

2375
2063
2125

O.K.
O.K.
O.K.

f 1

AA 2

7.100
6.980
7.350

21.280
23.500
19.430

71.470
69.100
73.140

3.12
3.11
3.12

138
200
200

1250
1625
1688

O.K.
O.K.
O.K.

1
BB 2

6.160
7.000
6.440

18.110
19.730
19.270

74.670
73.250
74.230

8
8

8H
9ki

3.14
3.13
3.13

125
188
200

1250
1500
1025

O.K.
O.K.
0. K

CC 2

6.600
6.350
6.380

21.330
18.650
22.140

71.980
74.800
71.450

3.09
3.12
3.12

125
125
125

1437
1250
875

O.K.
O.K.
O.K.

DD 1 2

4.950
5.110
5.000

17.360
18.550
16.960

77.650
76.220
77.960

6
6
7^

7
7

3.11
3.14
3.11

125
160
126

1813
1188
1000

O.K.
O.K.
0. K

EE 2

5.050
4.460
4.660

18.150
17.670
20.350

76.700
77.810
74.980

3.12
3.12
3.07

126
126
138

1937
1250
1400

O.K.
O.K.
O.K.

FFJI

4.440
4.730
6.000

17.450
19.480
19.050

78.070
75.770
75.920

3.11
3.09
3.11

150
138
138

1400
1437
1313

O.K.
O. K.
O.K.

f 1

GG 2
3

5.000
5.250
5.430

19.500
20.650
22.260

74.860
73.700
72.050

3M
3K

3.11
3.09
3.12

150
150
163

1437
1625
1563

O.K.
O.K.
O. K.

HH 2

4.960
4.850
4.950

21.750
19.050
19.020

73.030
76.000
75.900

2K
35i

3K
3K
4>i

3.14
3.07
3.12

150
150
138

1437
1500
1126

O.K.
O.K.
O.K.

' 1

•II 2

^ 3

5.140
6.010
5.800

19.080
20.520
21.310

75.650
74.380
72.830

3.12
3.11
3.14

125
88
126

1188
1063
1375

O.K.
O.K.
O.K.

f 1
JJ 2

7.420

21.920

70.420

3.12

125
150
160

1437
1538
1563

O.K.
O K

7.640

23.660

68.650

3.08

O.K.

f 1

KK \ 2

10.230
10.290
9.500

23.930
23.500
24.800

65.600
66.000
65.500

3.12
3.16
3.16

188
160
138

1376
1313
1275

O.K.
O.K.
O.K.

f 1

LL 2

7.030
8.200
6.300

22.260
23.900
20.370

70.600
67.900
73.230

6
6
6

8
8
7H

3.12
3.12
3.11

188
150
138

1663
1500
1625

O.K.
O.K.
O.K.

MM 1 2

9.350
10.320
10.520

6.850
6.850
7.120

22.500
23.930
23.610

27.170
25.900
23.230

68.000
65.580
65.700

65.740
67.000
69.480

6
6

3.15

126
138
188

188
163
150

1313
1313
1376

1126
1437
1376

O.K.
O. K.

3.14

3.062
3.107
3.137

O K

NN ■ 2
3

7
7
7

f 1

OO 2

6.750
7.660
6.980

23.450
24.260
22.930

69.050
67.800
70.000

8K>

;;;;;;;

3.137
3.122
3.137

126
100
126

1250
1313
1250

f 1

7.370
7.020
7.050

22.830
21.450
21.800

69.600 {
71.300 !
71.040

9
9
7

75
38
38

1125
1260
1376

PP 2

Digitized by

Google

512 Bayles on Inspection and Testing of Concrete.

Table I. — Tests of Cement and 1 : 2 Mortar. (Continued.)

Car

Number.

Percentage of Cement

Retained., on Sieve.

Setting.

Through initial

100 Mesh. 200 Mesh. 200 Mesh., Hra.

6.700
6.670
6.860

6.000
7.100
4.840

Final.
Hrs.

Specific
Gravity

22.570 70.450

21.180 I 72.960

21.900 I 70.900

17.750 76.800

24.100 68.680

18.760 76.280

7.800 23.320
7.660 22.460
7.300 ' 22.160

68.630 8
69.730 9
70.440 ' 9

Crushing

Strength

of 4-in. cubes,

inlb. persq.in.

24 hrs. 7 days.

125
125
150

150
113

1626
1688
1626

1750
1876

125 I 1876

3 122
3.107

163
163
150
150

1375
1375
2000
1875

Boiling
5 hours.

and the strength which can be safely counted on in construction
is rather above than below the average of the tests made.

requirements for materials and concrete.

The values given in Tables III and IV are not necessarily
true of all concrete, and perhaps it would not be amiss to describe
in some detail how this concrete was made.

In order that good concrete may be produced the first
requirement is that the specifications be correct, full, exact and
clear. This enables the inspector to do his work on a definite
plan without the annoyance of possible objections or interference
by the contractor's superintendent, who is sometimes more
interested in the quantity than the quality of the product. With
this object in view the specifications were made for this work.
Following are the items appl>'ing to the materials and manufac-
ture of concrete:

Concrete will be composed of one (1) part cement, two (2) part^ sand and
four (4) parts stone, with sufficient water to make a wet mixture. The
cement, sand and stone will be introduced into a batch mixer of approved
design and the mixer given 3 turns before the water is added. After the wat^r
is added the mixer will be revolved at least 12 times and to the satisfaction
of the engineer before the batch is dumped. Concrete must be placed immedi-
ately after mixing and well spaded to insure a dense concrete.

Cement, where used in these specifications, will mean Portland cement
of such quality as to meet the standard specifications of the American Society
for Testing Materials, and stand the tests pre^scribed in the rules and regula-
tions of the Bureau of Buildings, Borough of Brooklyn, City of New York.

Digitized by

Google

Bayles on Inspection and Testing of Concrete. 513

Table II. — ^Tests of Cement and 1 : 2 Mortar.

ar
nber.

Percentage of C

ement

Through
200 Mesh.

Setting.

Initial. 1 Final.
Hrs. 1 Hrs.

Specific
Gravity

Crushing

Strength

of 4-in. cubes,

in lb. per sq. in.

Boiling
5 hours

Nut

Retained
100 Mesh.

on Sieve.
200Meah.

24 hrs.

7 days.

5.93

6.28

18.97
19.15
20.20

74.90
74.70
73.50

3
3
2»i

3.152
3.122
3.137

450
563
469

1938
2375
2125

O.K.
O.K.
0. K.

f 1
2
3

6.36
5.75
6.

19.26
18.41
18.71

74.26
75.75
75.22

2}4

3.152
3.152
3.137

750
844
875

2313
2188
2563

0. K.
O.K.
O.K.

f 1
2
3

6.7

7.26

6.75

19.85
19.90
19.90

73.25
72.70
73.22

1».4

2»yi

2»4

3.122
3.137
3.137

688
513
594

2125
1750
1813

O.K.
0. K.
O.K.

f 1
2
3

6.14
5.04
6.09

20.21
19.82
15.65

73.43
74.01
77.94

2
2

2?4

3>i

3.137
3.16S
3.152

406
406
406

2000
1938
2063

O.K.
O.K.
O.K.

6.32

6.5

6.17

20 02
18.20
19.21

73.51
75.21
74.48

2
2
2

3.152
3.122
3.122

313
469
388

1875
19.38
2000

O.K.
O.K.
O.K.

ll
13

5.38
6.67
6.41

20.02
16.98
19.36

74.39
76.09
74.06

3.152
3.152
3.168

475

388
438

1875
1688
1938

O.K.
O.K.
O.K.

f 1
2
3

5.07
6.32
6.39

19.83
20.13
20.38

74.86
73.38
73.08

3»4

3.152
3.152
3.137

375
525
531

2125
2813
1813

O.K.
O.K.
O.K.

f 1
2
3

6.61
6.13
5.92

18.84
19.06
19.41

74.29
75.58
74.43

2'<i
2^

S14

3?4'

3'^i

3.137
3.152
3.152

.531
563
531

2188
2188
2000

O.K.
O.K.
0. K

f 1
2
3

6.17
5.06
6.18

18.92

19.09.

19.32

74.68
75.53
74.29

3«4

3.168
3.152
3.168

531
525
513

2125
2125

2188

O.K.
O.K.
0. K.

f 1
2
3

6.13
7.73
7.20

18.84
18.96
17.64

74.81
73.08
74.93

4'2

3.137
3.16K
3.137

500
625
563

2125
2250
2250

O.K.
O.K.
O.K.

f 1

{ 2

6.17
5.81
6.13

17.86
19.07
19.20

75.61
74.89
74.38

3
2^

4>i

3.152
3.152
3.168

438
388
538

2125
2188
2000

O.K.
O.K.
O.K.

f 1
2
3

5.86
5.12
6.09

18.42
18.63
19.13

75.41
75.93
74.59

3.137
3.168
3.168

450
531
325

2125
2063
2000

O.K.
O.K.
O.K.

( 1

, 3

5.16
6.03
5.82

19.13
18.79
19.03

75.38
74.92
74.97

4
4M

3.168
3.152
3.152

613

688
5.'>0

2313
2375
2063

O.K.
O.K.
O.K.

f 1
2
3

5.20
5.43
6.12

19.41
19.21
19.02

75.13
75.19
74.70

3^
3H

4j.i

3.152
3.137
3.152

525
600
538

2125
2438
2188

O.K.
O.K.
O.K.

f 1
2
3

6.12
5.87
6.02

18.80
19.21

18.84

74.91
74.79
75.01

35i

434

4^4

3.168
3.168
3.152

563
550
775

2125
1875
2313

O.K.
O.K.
O.K.

f 1
2
3

5.82
6.45
8.24

19.63
19.47
17.14

74.38
73.43
74.14

5
5
6

3.168
3.168
3.137

594
525
525

2250
1938
1938

O.K.
O.K.
O.K.

f 1

7.42
9.76
7.40

17.39
19.31
18.39

75.02
70.80
74.03

3>i

3.152
3.152
3.137

313
250
450

2125
1875
2250

O.K.
O.K.
O.K.

Digitized by

Google

514 Bayles on Inspection and Testing of Concrete.

rABLE 11

Percei
Retained
100 Mesh.

6.47
9.50
8.96

.— Tests

of Cemc

sment

Through
200Me8h.

nt and

Sett

Initial.
Hrs.

4
3^

1 :2\

ing.

Final.
Hrs.

lortar.

Specific
Gravity

3.137
3.168
3.152

{Conlinued)

Crushing
Strength

of 4-in. cubes.

in lb. per sq. in.

Car

Number.

atage of d
on Sieve.
200 Mesh.

Boiling
5 hours.

24 hrs.

300
300
219

7 days.

R 2

18.33
16.24
16.88

75.02
74.08
74.05

2250
1938
1875

O.K.
O.K.
0. K.

[ 1

S 2

, 3

8.66
9.34
8.12

17.20
16.70
16.91

74.02
73.81
74.83

3' '2

4'a

4*4

5' 2

3.168
3.168
3.152

300
225
219

1625
1750
2000

O.K.
0. K.
O.K.

T 2

6.79
7.04

8.48

18.62
17.47
16.60

74.41
75.30
74.81

5'4

6>i
5H

6»4

6'i
6>i

3.137
3.122
3.168

125
100
113

1875
1625
1875

O.K.
O.K.
O.K.

U { 2

8.43
5.87
7.70

17.05
18.92
17.84

74.37
74.97
74.29

5»4

5^4

7*4

3.152
3.152
3.168

113
106
88

1938
2188
1688

O.K.
0. K.
O.K.

V 1 2

6.88
7.13
6.84

17.92
18.22
18.39

74.94
74.47
74.51

4j;

3*4

6
5
5

3.122
3.152
3.152

475
438
438

1963
1875
2000

O.K.
O.K.
O.K.

W 1 2
13

7.31

8.48
8.82

18.06
14.70
16.93

74.48
76.59
74.97

4H
4h

3.122
3.162
3.152

419
425
288

2125
2313
2063

O.K.
0. K.
O.K.

9.03

7.14
8.23

16.17
17.09
16.18

74.62
75.49
75.44

4 4

4H
4H

3.137
3.168
3.168

213
181
150

1938
1875
1688

O.K.
O.K.
O.K.

Y 11

7.13
8.08
8.32

17.95
16.69
18.17

74.79
75.06
73.29

4%
4h
4H

3.152
3.137
3.122

169 1 1750
325 1 2188
294 2000

O.K.
O.K.
O.K.

f 1

Z 2

7.93
8.12
6.92

18.31
18.22
18.05

73.49
73.43

74.88

4H
4».i
4,4

61 2

3.137
3.152
3.137

263
250
288

1625
1437
1563

O.K.
0. K.
O.K.

f 1

AA 2

8.72
7.88
9.47

18.39
18.27
17.62

72.74
73.60
72.79

6K
64
6»i

3.122
3.122
3.152

288
350
350

1563
1688
1437

O.K.
O.K.
O.K.

f 1

BB 2

11.40
9.13
8.72

18.70
18.99
18.04

69.78
71.64
73.07

3.152
3.137
3.137

363
344
213

1437
1500
1437

O.K.
O.K.
O.K.

CC 2

6.93
8.31
8.43

19.09
19.01
18.77

73.84
72.43
72.63

4H
4'i
434

3.137
3.107
3.122

219 1813«
238 I 2«>3«
282 1 1875>

O.K.
O.K.
O.K.

r 1

DD 2
3

8.70
7.93
8.41

17.34
18.04
18.57

73.84
73.71
72.83

4l'l
4H
4H

5H
54

3.137
3.137
3.122

338 ' 2063«
344 20002
350 1 2375»

O.K.
O.K.
O.K.

EE I2
l3

12.24
9.17
8.41

16.84
17.32
18.69

70.79
73.24
72.74

4U
6U'

3.152
3.152
3.137

275 2063
344 1 2188
313 2250

O.K.
O.K.
O.K.

FF 2

9.96
9.22
8.43

17.58
17.93
18.34

72.29
72.61
73.02

6
64

3.122
3.137
3.137

319 2000
244 1875
250 , ^000

O.K.
O.K.
O.K.

f 1

GG 2

10.92
13.30
11.76

18.63
18.95
18.72

70.12
67.50
69.06

5>4

3.184
3.122
3.122

175 1 1563
275 1375
100 1 1563

O.K.
O.K.
O.K.

HH J2
3

4.75
5.28
5.15

19.31
24.87
21.62

75.9

69.72

73.12

2
214

3)i

3.12

3.2

3.152

938
1063
1063

2500
2000
3063

O.K.
O.K.
O.K.

> Eight days

Digitized by

Google

Batles on Inspection and Testing of Concrete. 515

Car

f 1
2
3

f 1

f 1
2
3

f 1

f 1
2
3

1
2
3

f 1
2
3

Table II, — ^Tests of Cement and 1 : 2 Mortar. {Continued)

Setting.

Percentage of Cement

Retained on Sieve.
100 Mesh. 200 Mesh.

(1
II

5.36
7.82
6.47

4.75
5.46
5.83

5.00
5.13
5.09

6.64
6.23
6.06

5.18
6.05
6.13

5.97
6.13
5.99

7.10
5.94
6.12

5.68
6.32
6.29

6.13
6.07
5.84

7.02
6.87
6.92

4.53
4.60

4.87

7.03
6.61
6.72

5.87
5.69
6.12

6.35
5.86
5.92

5.88
7.44
6.78

5.97
6.31
6.41

6.74
6.37
7.01

22.28
26.06
24.63

25.93
26.97
24.66

22.72
21.87
22.36

23.01
23.18
21.61

23.64
24.16
24.42

23.19
22.94
21.82

21.78
22.31
22.14

25.04
22.76
24.07

22.77
22.69
21.57

23.62
20.93
22.17

22.05
20.66
21.33

23.65

23.115

22.95

22.06
23.12
22.36

24.13
23.70
23.15

19.26
23.01
23.55

23.82
23.77
21.95

20.65
22.06
20.85

Through
200Me8h.

72.19
65.96
68.79

69.16
67.44
69.39

72.16
72.89
72.47

70.24
70.47
72.20

71.06
69.63
69.30

70.71
70.85
72.11

71.01
71.61
71.60

69.15
70.79
69.51

71.01
71.20
72.28

69.20
72.08
70.80

73.05
74.45
73.61

69.20
69.80
70.13

71.91
71.03
71.40

68.70
70.21
70.49

74.45
69.53
69.41

70.03
69.61
71.39

72.40
71.29
71.98

2
2

IH
IH

2V^
2>i

2?4
2H

2>4

2H
2H
2H

Initial.! Final
Hrs. I Hrs.

2M
2H
2

2H.

2
2
2

Crushing
Strength |
Specific of 4-in. cubes, Boiling
Gravity! in lb. per sq. in. 5 hours.

O. K.
O.K.
O.K.

O.K.
O.K.
O.K.

Pigitized by

Google

516 Bayles on Inspection and Testing of Concrete.

Tablk II. — Tests of Cement anil 1 : 2 Mortar. (CorUinued)

Car

Number.

PercentaRp of Comont
Retained on Sieve. ^^^^^^^
lOOMeah. 2(M)MeHh. 2(K)Me.sh.

5.91 18.22 75.05
6.(19 15.42 ; 77.75
6.21 15.31 78.14

Sett
Initial.

h™.

Final.
HrH.

3
3

2»4

1 Speeific
Gravity

3.168
3. IS.?
3.122

Crushing

Strength

of 4-in. eubes.

in lb. persq. in.

Boiling
5 hours.

24 hrs.

938

688
625

7 day b.

r 1

ZZ 2
.3

2438
2250
2375

O.K.

. O. K.

O.K.

f 1
AB < 2

8.06
5 52
7.02

17 93
17 83
20.05

73.60
75.33
72.70

3
2V

3.152
3.152
3.152

563
375
375

1088
2000
1750

O.K.
O. K.
O.K.

f 1

AC \ 2

I 3

9.20
7.21
0.92

19.03

17.95

70.66
73.67
75

2
2
2

1 3.162
' 3.152
, 3.137

1250
1000
1250

2375
2250
2875

O.K.
O.K.
O.K.

f 1

AD 2

> 3

8.35

7.47

19.35
19.80
21.02

72.12
73.21
71.51

3 4
33 a

! 3.137
3.137
3.137

1125
1188
1063

2125
2063
2438

O.K.
O.K.
O. K.

f 1
AE 2

6.84

6.30

19.35

19.27

73.60
72.89
74.35

3.137

1 3.137

813^
938
938

2375'
2125
2250

O.K.
O.K.
O.K.

AF 1 2

6.44
6.55
6.9

18.45
18.5<i
18.8

74.83

74.07

3
3

3.152
3.152
3.137

813
600
438

1750
2188
2313

O.K.
O.K.
O.K.

f 1

AG 2

6.7

6.12

7.13

18.71
19.00
19.28

74.55
74.61
73.40

3'4

3
3

4
4
4

3.137

j 3.152

688
625
781

1750
1688
1750

O.K.
O. K.
O.K.

f 1

AH 2

6.23
5.47
7.11

19.66
19.80
18.80

74.05
74.58
73.87

2»4
2«4

4M
4

3.152

1 3.137

3.137

656
875

781

2250
2250
2438

O.K.
O.K.
O.K.

AI 1 2
3

7.92
8.46
8.17

20.08
21.19
20. .58

71.80

70.71

3'..

4li

1'^

3.137
3.137
3.137

656
719
663

2125
2063
2250

O.K.
O.K.
O.K.

AJ 1 2

6.77

5.76

20.47
21.21
19.58

72.32
71.80
74.53

1«

3.137
3.137
3.137

531
503
594

2000
2063
2188

O.K.
O.K.
O.K.

AK 2

6.09
6.16
6.04

19.66
19.98
19.82

73.74

74.09

2H
2H

3
3
3

3.152

. 3.137

3.152

500
375
594

1938
1625
2250

O.K.
O.K.
O.K.

f 1
AL { 2

5
5
5.25

19.45
20.21
20.19

75.35
74.70
74.24

3
3j.i

: 3.168

3.137

, 3.152

531
600
406

2188
2313
168S

O.K.
O. K.
O.K.

f 1

AM < 2

0.82

7.37

19.97
18.20
18.42

73.63

74.03

2\i
2'.i
2U

3.122
3.137
3.137

638
663
513

2188
2188
1813

O.K.
O.K.
O.K.

f 1

AN 2

6.82
6.53
6 41

19.75

19.90

72.80
73.45
73.45

2'i

2'2

3.152

1 3.137

388
531
375

1750
2000
1813

O.K.
O.K.
O.K.

f I

AG 2

, 3

5.95
6.25
6.42

19.40
19.05
20.11

74.50
74.60
73.35

1?4

3
3

2?4

3.122

3.137

1 3.137

475

469
481

2750
2250
1813

O.K.
O.K.
O.K.

» Sand contained 58 per cent loam.

Digitized by

Google

Bayles on Inspection and Testing op Concrete. 517

Table III. — Crushing Strength of 4-in. Concrete Cubes.

Strength in

Str'gth

Arc

days.

336

Strength in

Str'gth

Age

days.

Date made.

lb.
24 hr.

375

per sq. in.
7dy. ,28dy.

1250' 2425

in
lb. per
sq. in.

3688

Date made.
_
11-17-10

lb.
24 hr.

250

per sq. m.

7 dy. !28dy.

■ — 1

2125* 35006

in
lb. per
sq. in.

8-18-10

4688

240

8-18-10

375 .... 2475

3566

336

11-18-10

188

1750 2750^

5000

241

8-19-10

500 1375 2500

4088

335

11-19-10

313

1250 4250

6260

252

8-19-10

375

19001 2188

3750

335

11-21-10

250

1500 3375

4375

238

8-20-10

500

1125 2875

4250

334

11-29-10

275

1875 1 3500

4750

230

8-22-10

500

1250 2225

4500

331

1 11-30-10

188

1250 3726

4875

226

8-25-10

625

2000 , 2875

4375

329

12- 1-10

200

1875 ' 3125

4375

224

8-25-10

500

2000 , 3938

5626

328

! 12- 1-10

200

1875 4938

6750

239

8-26^10

375

1350 ' 2875

4375

327

12- 2-10

200

1875 : 3188

4625

223

8-26-10

375

1750 1 2875

4250

322

12- 5-10

263

1250 ; 2750

4875

225

8-30-10

375

875 3000

4376

323

12- 8-10

188 1500 1 3188

5000

221

8-31-10

750 ' 2275 i 3500

4625

321

12-15-10

200 1 2250 3438

4750

214

9- 6-10

875 , 1975

3500

5000

316

12-19-10

200 1 2375*' 4438

&- 7-10

438

1688

2500

3875

316

12-20-10

350 ; 1900 2938

4125

izi 2 ■

9- 8-10

563

1500

3125

4375

313

12-27-10

250 1625 3250

4875

205

9-13-10

563 1375 3125

4813

308

12-28-10

313 2375 402.'^«

6125

203

9-14-10

500 1875 3375

4626

307

12-28-10

2.W

1875 1 418^"

4688

201

9-14-10

438 1625 3250

5063

307

12-29-10

438

3026 ! 5563

6125

200

9-15-10

750

1875 4375

6250

307

1 12-30-10

275

2750 5500

5500

200

9-15-10

625

2625 3813

4813

306

1 1- 9-n

188

1600 3063

4625

189

9-16-10

563

2250 3750

5563

306

1-11-11

150 1 1400 , 2938

3125

190

9-16-10

600 2188 3750

5000

305

1 1-12-11

400 3150 1 4000

7500

197

9-17-10

375 1500 3125

5250

304

1 1-12-11

250

1688 3750

5000

197

9-19-10

500 1938 2500

5250

303

; 1-14-11

250

1625 1 3250

4875

195

9-20-10

688 , 2250 3500

4500

302

1-19-11

225

1438 3125

4875

190

9-20-10

813 2250 3938

4500

301

1-19-11

313

1688 3125

4500

190

9-21-10

725 1875 3500

4938

300

1 1-21-11

150

1313 2625

3625

188

9-21-10

563

2125 ! 37 .fin

4760

301

1-23-1 1

125

1250 2025

9-22-10

563

2500

3600

5875

294

: 1-23-11

150

1313 i 2500

4250

im '

9-22-10

500

1875

2938

4375

299

1-24-11

338

1625 2625

3750

185

9-23-10

750

2250

2688

5376

298

1-26-11

262

1250 2375

4125

168

9-23-10

438
500

2000
2250

3125
3125

1-26-11
1 3- 8-11

438»
188

1250 2250
1625 j 3250

3500
4750

183

9-24-10

' 4688 '

299

124

9-26-10

1000

1750

2500

4688

297

! 3- 8-11

200

1750 1 287.-)

4376

124

9-27-10

600 1875

3500

4188

294

3- 9-11

175

1938 2875

9-27-10

875 1875

3125

4626

296

3- 9-11

225

1975 , 3938

■ '4875 ■

124

9-28-10

563

1575

2750

3750

294

3-10-11

188 ■ 1876

3750

5813

132

9-29-10

500

2063

3000

4260

292

3-10-11

263 2375

5625

121

9-30-10

563

1875

2875

4125

293

1 3-11-11

250

2500

5000

7188

123

10- 3-10

2.'>0

1250

2000

3125

288

1 3-13-11

313

2188

4813

65(K)

121

10- 4-10

500

2125

3125

4938

289

3-14-11

2.50 2750

4625

6750

128

lO-U-lO

500

2000

2750

3500

279

3-14-11

188 2125,4188

6250

118

10-13-10

375

1875

2813

4313

277

3-18-11

188

2250 5500

6125

116

10-14-10

500

1750

2750

4250

276

! 3 20-11

350

2500 ! 4750

7125

114

10-15-10

375

1875

2250

3563

273

1 3-20-11

288 2500 4375

5625

114

10-17-10

500 2250

2938

4813

274

3-21-11

375 3(X)0 i 4750

6625

113

10-19-10

500

1250 2000

4375

268

3-21-11

275 2500 5000

6875

112

10-19-10

688

1875 2750

4750

269

3-23-11

313 2375 1 5000

5750

110

10-25-10

313

1250 ! 2250

3625

261

3-30-11

250

2.500 1 4625

6750

103

10-25-10

313

1375 ; 2260

3126

262

3-31-11

150

2000 1 4875

6260

101

10-26-10

375

2063 2688

4625

266

4- 1-11

313

2563 5250

7600

102

10-28-10

350

1688 2938

4876

264

4- 3-11

250

2500 . 4625

6126

100

1O-31-10

250

1563 2000

4375

256

4- 3-11

250

2625 5000

5125

100

10-31-10

263

1250 2750

4750

256

4- 5-11

250

2125 4625

6000

10-31-10

313
313

2125
1750

3125
2750

4- 6-11
4-13-11

375
375

2.563
3125

4500
4625

5125
6625

11- 1-10

4938

i260

11- 2-10

500

1250

2875

6875

251

4-13-11

438

3125

4750

6126

11- 2-10

725

2125

2875

4003

2.58

4-14 11

375

2750 4625

5625

11- 7-10

350

2438

4600

6500

250

4-14-11

43810

3375-^ 5625

4750

11- 7-10

313 2250 4125

6063

255

4-15-11

563"

2250 3500

4375

11- 9-10

313 2250 3125

5500

246

4-18-11

375

1625 3600

4625

11-10-10

350

1625 3125

4750

247

4-22-11

260

1625 1 3375

4063

11-10-10

375

1500

3000

6000

246

4-25-11

375

1875 3750

4750

11-11-10

250

2000

3125

4125

245

4-27-11

438

2375 ' 4375

6125

11-11-10

188

2000

3125

5125

2.>0

4-28-11

750 2500 5000

6875

11-12-10

250

2125

27505

5875

248

5 8-11

500 2250 4125

5000

11-14-10

188

1375

2375

4250

243

5-12-11

626 1875 4375

5000

11-15-10

225

1750

3813

5875

246

5-18-11

625 2250 4.375

5375

11-16-10

188

1250

2500

4188

241

5-20-11

875 1750 3625

4063

11-16-10

313

2125*

3125

1 14 days

<12

days.

» 27 daj

^8. W

21 hours

» 30 hour

B, •8 c

lays.

•29 da]

fS, »

53 hours

> 30 days

•70

days

» 43 hoi

ITS.

r~>^

Dig

tized by

518 Baylbs on Inspection and Testing of Concrete.

Table IV. — Crushing Strength of 4-in. Concrete Cubes.

Date made.

7-15-11
7-17-11
7-lS-n
7-18 11
7-19-U
7-24-H
7-25~n
7-25-11
7-afl-ll
7-2fl-ll
7-31-n
8- J-H
ft- J-il
8- 2-11
8- 2-il
S- 3-11

8- 3-n

8- 4-11

8- 4-11
8^- 8-11

8- li-ll
S'l&-]1
8-11-11
S-ll-ll
8-H-ll
SI- II 11
8-lt>-ll
8-18-11
8-25-11
8-28-11
8-28-11

9- 1-11
9- 1-11
9- 2-11
9- 5-11
9- 7-11
9- 8-11
9-12-11
9-13-11
9-14-11
9-16-11
9-18-11
9-18-11
9-19-11
9-20-11
9-20-1 1
9 21-11
9-22-11

IK p{!r b(}k In,
!J4hr 7dy. l2Rdy

lOOOJi;
50;t
08^

75LJ

esH
7at3

5m
m:i

^35 I
62A I

essM

BOtI
625

as8

525

sea
Si:i
5fla

635

m>»
56:*
fi2V

4;^9 I

563 !

563

625

563

375
1250*

625

500

375

563

375

500

9387

325

450

487

413

438

438
1500»

1500 |2lS8
1750 1^875

1750 3I8S

1500 2rm

I37f» 2375

11125 2S75

1375'* 2813

ViJii ims

rim '2ms

12505 2im
15(XI^ 22r»o
i:j7/i '.utm^
m:i 2i:iji

125C] 2750
llStH if500
150tl 287,'i

Umii 2125

iOf>:i iihH
H7:j ntiK

li;S5* 2260
1188 2.563
1313 2313
1125 2188
1000 ,1813
1375 '3188
1250 2(525
1188 2813
1250 12500
1125 ,2625
875 2063
1063 1 2688
1063« '2563
1063 J2438
1375 2563
1000 12250
1188 2250
1188 2375
1250* 1938
1500» 2250
15(K) 2563
1688 12875

in I Age
lb. per , ,»°
aq. in. I days.

I i

I 116» I

I 177 I

, 115» I

! 176

I 109» I

' 170 I

I 170

1 107»

I 169 I
102>,,

Date made.

Strength in
lb. per aq. in

,24 hr. 7 dy. '28dy.| aq. in.

3625
3950
4875
4519
5000
4031
5063
4000
4313
4688
4269

I 2938

3813

1 3875

4069

4519

5125

4169

I 4288

I 5625

, 3625

3637

! 3875

3719

I .3875

I 3438

, 4063

I 3788

3750

3138

4813

I 4.375

3938

4625

I 3688

3875

5000

4125

4375

4.500

39,38

I 3813

I 42.')0

34.38

4063

4250

4875

38 1
1001
100»

991

99»
*98»
<98« I

94»
155
154

91»
1.53

881 I
1.50

861 I
146

771

1.36

136

1.32

132

131

128

126

125

121

120

119

117

115 I

115

114 i

113

112

111

^ Broken at Columbia University.

* 6 days, 7 hours.

* 5 day^, 2 hours.

* Contained 6 per cent loam, extra 10 per
oent cement used.

* 5 days.

* 3 days.

' 53 hours.

9-25-11
9-25-11
9-27-11
9-28-11
9-28-11
10- 2-11
10- 2-11
lO- 3-11
10- 3-11
10- 5-11

10- 6-11
10-10-11
10-10-11
10-12-11
10-12-11
10-16-11
10-17-11
10-17-11
10-19-11
10-20-11
10-23-11
10-24-11
10-25-11
10-26-11
10-30-11
10-31-11

11- 1-11
11- 2-11
11- 2-11
11- 3-11
11- 4-11

11- 6-11
11-11-11
11-1.3-11
11-13-11
11-14-11
11-14-11
11-17-11
11-20 11
11-20-11
11-21 11
11-22-11
11-24-11
11-28-11
11 2t)-ll
11-29-11

12- 1-11
12- 1-11

719

6.50

688

388

344

263

294

469

438

419

4.38

438

388
i 375
I 375

406

375

450

325

294 I

300

.344

263 1

263

300 .

238 I

113">

lOO"
I ..",

250"

400»i
I 450";
I 94 '

169

2.50 1

275 I

125

225 I

194 •

331

100

175

200
•'188>a
"156"!
, 219

156

2250
2313
24.38
2000
2188
2313

1375
1313
1625*
1000
875
9.38
875 ;2063
1188 2563
1125« 12063
1000 2250
1063 |2813
1063 2188«
1188«;18757|
1063 I2OOO
875 11938 I
1000 2063 I
969 12313
1.500« ,2438
788 '2125
825 11875
788 I2I88 '
919 2063 I
800» 11813 '
675 17,50 I
888 12875 |

969 12563
600t 1 2125
469 12063 I
400 i2188 I
6.56»|l625 1
600 12438 ,
781 ;i625 I
663 I2OOO
4,38 11625
588 19.38 I

713

788
875>
800
888
1125

1625

19.38

1813 I

1625

1813

2000

813 '2000
7.50 13063
1000 1938
688 ,1688
875u!l813
719 11,563
869 ,1813

4500
4000
4625
3625
3625
3813
3625
4375
3688
3625
4250
4250
3563
3688
3875
4125
4313
4188
4063
3375
3438
3875
3688
3375
5125
5063
3625
3875
3875
3625
3438
3438
2875
3250
3000
3000
3250
3125
2813
3125
3125
.3000
4438
3063
2500
2188
2625
3250

Age

in
dasrs.

108
108
106
105
105
101
101
100
100
98
97
93
93
91
91
87

84
83
80
79
78
77
73
72
71
70
70
69
68
66
61
59
59
58
58
55
52
52
51
50
48
44
43
43
41
41

•6 days.

•29 days.
" Cube not set.

11 Cube broke when removed from forms.
" 48 hours.
" 9 days.
»« 4 days.
» 7 days.

Digitized by

Google

Bayles on Inspection and TESTnjra op Concrete. 519

All sand must be washed, clean, sharp, silicious and free from injm'ious
matter.

Ail stone shall be clean crushed trap rock, free from dust and to contain
no particle that will not pass through a f-in. ring.

There was little diflSculty in getting satisfactory materials.
Standard brands of Portland cement were used. The sand was
what is known in New York as "Cow Bay," and was satisfactory
with a few exceptions when, for some reason, it was impossible to
get washed sand from the usual source and other sands were tried
temporarily to take its place in order not to delay the work.
Some of this sand was satisfactory, some was condemned out-
right and one scow load was accepted on condition that lO per
cent additional cement be used, — a doubtful expedient. The stone
was a Hudson Palisades trap, commercial J-in. size and proved
generally acceptable.

operation of plant.

The mixer, a batch mixer, was set in a pit so that the top of
the receiving hopper was a little above the floor of the wharf.
The sand and stone were measured approximately in wheel-
barrows or buggies and the cement by the bag, reckoning a bag of
cement at 0.95 cu. ft. The aggregates were all put into the
hopper, the trap opened and the batch pushed into the mixer.
As soon as all the aggregates were clear of the hopper the water
was added, the mixer revolved 12 times and dumped, while
another batch was being prepared.

At first the water was dipped with a pail from a barrel on the
platform and splashed through the hopper into the mixer. This
kept the sides of the hopper always wet, preventing the easy dis-
charge of the dry aggregates and also delayed the work by using
the hopper as a funnel while another batch should have been in
course of preparation. These were matters primarily for the
contractor's attention, but this method produced concrete of such
varying degrees of consistency that the plan shown in Fig. 2 was
suggested to the contractor and immediately installed. By this
method the water flows constantly or nearly so from a f-in. pipe
into the upper barrel; from this barrel a l|-in. pipe with quick
opening lever gate valve leads to the lower barrel. From the
bottom of the lower barrel a 3-in. pipe, also fitted with quick

Digitized by

Google

520 Bayles on Inspection and Testing of Concrete.

opening valve, leads into the mixer under the hopper trap. A
number of small holes were bored in the side of the lower barrel.
By trial the right amount of water for a batch of concrete was
determined and all the holes below the line corresponding to that
amount plugged. Only the inspector at the mixer was allowed to
vary this amount. After all the dry material had run into the
mixer, which it did more readily since the hopper was always dry,
the hopper trap was closed and the 3-in. valve opened. Only a
few seconds was required to empty the barrel, when the 3-in.
valve was closed, the IJ-in. valve opened and the water allowed

FIG. 2. — ARRANGEMENT OF WATER SUPPLY FOR MIXER.

to run until it had filled the lower barrel to the holes in the side>
which served both to call the attendant's attention and to drain
off surplus water. By this method a satisfactory mixture was
achieved and the time of mixing reduced by at least a third.

The concrete as thus produced was of such consistency that
it was wheeled to the work — at most 300 ft. — ^without any
apparent segregation of material and when the buggy was tipped
the whole mass contained slipped, rather than poured, into the
forms, leaving the buggy clean. The diflference between concrete
that is just wet enough and concrete that is too wet is very slight

Digitized by

Google

Bayles on Inspection and Testing of Concrete. 521

and only experience can determine what is right and constant
care maintain it.

The batch was dumped from the mixer into a bucket and
immediately hoisted and dumped into a hopper above the floor
where it was to be used. From this hopper it was drawn off into
concrete buggies and wheeled over plank runways to the work.
The buggies had a capacity of 6 cu. ft., but in practice each man
took about two thirds that much.

The columns were poured first, preferably 24 hours before the
slab. The concrete in the columns was puddled with long
wooden bars about 2 x 3 in., which were also used to hold the
reinforcement in place as the concrete was being poured. Every
buggyf ul of concrete was thoroughly puddled before the next was
added.

In concreting the floors it was the general practice to fill the
beams and girders first, allowing the concrete to spread on to the
adjoining slab rather than from the slab into the beams and
girders. The floor concrete was puddled with narrow wooden-
handled iron spades. Care was taken that no concrete stood
long enough, to form a joint before new concrete was added.
When construction joints were necessary they were made at the
middle of the span and were vertical. Before new concrete was
added the surface of the joint was roughened and washed with
neat cement grout. The concreting gang was followed closely
by the finishers who screeded the slab to the required thickness
and troweled the surface to a smooth, hard finish.

While the inspection was close and strict the contractors
co-operated heartily and the work was done without friction.

Digitized by

Google

COMPARATIVE TESTS OF THE STRENGTH OF CON-
CRETE IN THE LABORATORY AND IN THE
FIELD.

By Rudolph J. Wig.*

It is common practice at the present time to define a desired
strength of concrete by stating that it shall be made of good
quality materials in certain given proportions. Assuming the
use of average concrete material a 1:2:4 proportion mixture is
considered as developing an ultimate compressive strength of
2000 to 3000 lb. per sq. in. in 28 days, 3000 to 3500 lb. per sq.
in. in 13 weeks, etc. These values are based upon the results of
many thousands of laboratory tests which imdoubtedly represent
the true values which can and should be obtained with good
materials if properly used in the field. Much attention is given
to careful selection and the testing of the materials entering into
the concrete and usually in case of the failure of the. concrete, the
cause of failure is attributed to the cement or poor aggregate.

Several years ago an investigation was made in the Struc-
tural Materials Testing Laboratories of the United States Geolog-
ical Survey at St. Ix)uis, Mo., to determine the influence of dif-
ference in workmanship upon the strength of reinforced concrete
in flexure. In connection with this investigation compression
tests also were made on the plain concrete molded into 8 x 16 in.
cylindrical t«st pieces. The work of three prominent contractors
of the city of St. Louis was compared and the results of the tests
are of much interest.

The materials, cement, sand and aggregate, were carefully
weighed and proportioned under the supervision of the laboratory
force which ensured each contractor receiving exactly the same
amount of material so that the variation in the results of the
tests of the concrete prepared by the contractors was due entirely
to a difference in the method of mixing and handling the mate-
rials.

In the beam tests for hand-mixed concrete in the proportion

* Assistant Engineer, Bureau of Standards. Washington, D. C.

(522)

Digitized by

Google

Wig on Tests of Strength op Concrete. 523

1 part cement to 3 parts sand to 6 parts stone when 4 weeks old,
there is a difference of over 200 per cent in the maximum load
carried by beams made by different contractors with the same
materials. The total maximum load varied from 4850 to 16,310
lb. For the machine-mixed concrete the variation is about
90 per cent ranging from 8160 to 14,500 lb. total maximum load.
The quality of the concrete can perhaps be better observed from
the compression tests on the cylinders which were molded from
the same mixtures as the beams. The placing of the material in
the cylinder molds, however, was done by the laboratory force
so that the difference in the results of these tests may be entirely
attributed to the difference in the method and the thoroughness
of the mixing and the difference in the consistency of the concrete
as mixed, as each contractor was instructed to prepare a concrete
such as he used in practice and to exercise no more care in its
preparation than he would under working conditions.

This privilege of varying the quantity of water perhaps had
the greatest effect upon the strength although it was also mate-
rially influenced by the method of mixing as is shown in the
results where two of the contractors used the same percentage of
water in mixing.

A summary of the results is as follows:

1:3:6 Gravel Concrete, Machine Mixed. (Average of 3 test pieces.)
Compressive Strength in lb. i)er sq. in.

4 weeks. 13 weeks.

Max. Min. Ave. Max. Min. Ave.

Company A 998 650 844 1470 1180 1363

B 577 513 549 1020 878 944

C 725 623 679 1243 956 1138

Laboratory 1074 961 1032 1653 1540 1611

There is an approximate difference of 100 per cent in individual values
which range from 513 to 1074 lb. per sq. in. at 4 weeks and from 878 to 1653
lb. per sq. in. at 13 weeks.

1:3:6 Gravel Concrete, Hand Mixed. (Average of 6 test pieces.)
Compressive Strength in lb. per sq. in.

4 weeks. 13 w^ceks.

Max. Min. Ave. Max. Min. Ave.

Company A 988 510 746 1030 561 807

B 672 409 518 590 522 568

C 640 321 475 881 640 764

Digitized by

Google

524 Wig on Tests of Strength of Concrete.

There is an approximate difFerence of 200 per cent in individual values
ranging from 321 to 988 lb. per sq. in. at 4 weeks and an approximate dif-
ference of 100 per cent at 13 weeks, the values ranging from 522 to 1030 lb.
per sq. in.

1:3:6 Limestone Concrete, Hand Mixed. (Average of 6 test pieces.)
Compressive Strength in lb. per sq. in.

4 weeks. 13 weeks.

Max. Min. Ave. Max. Min. Ave.

Company A 447 398 422 810 668 745

B 405 364 390 584 522 566

C 999 507 693 1380 575 864

There is an approximate difference of 200 per cent in individual values
ranging from 364 to 999 lb. per sq. in. at 4 weeks, and an approximate dif-
ference of 150 per cent at 13 weeks, the values ranging from 522 to 1380
lb. per sq. in.

1:2:4 Gravel Concrete, Machine Mixed. (Average of 3 test pieces.)
Compressive .Strength in lb. per sq. in.

4 weeks. 13 weeks.

Max. Min. Ave. Max. Min. Ave.

Company A 1331 1211 1265 1834 1675 1740

B 1741 1461 1630 2195 2105 2157

C 2375 1991 2216 2290 1958 2132

Laboratory 2589 2540 2572 2698 2634 2672

(All test pieces of this series were exposed to the weather.)

There is an approximate difference of 100 per cent in individual values

ranging from 1211 to 2589 lb. per sq. in. at 4 weeks and of 70 per cent at

13 weeks values ranging from 1675 to 2698 lb. per sq. in.

1:2:4 Gravel Concrete, Machine Mixed. (Average of 3 test pieces.)
Comprej^sivo Strength in lb. per sq. in.

4 weeks. 13 weeks.

Company A 1650 1160 1443 1890

B 1942 1691 1787 2607 2294 2485

C 1966 1615 1828 2389 2220 2308

Laboratory 2487 2089 2312 2899 2735 2809

(All test pieces of this series were cured in the laboratory.)

There is an approximate difference of 130 per cent in individual values
which range from 1160 to 24vS7 lb. per sq. in. at 4 weeks and of 50 per cent
at 13 weeks, values ranging from 1890 to 2899 lb. per sq. in.

The yield point of the concrete in the last two series varied
from 400 to 1000 lb. per sq. in.; thus in some cases the yield point

Digitized by

Google

Wig on Tests of Strength of Concrete. 525

of the concrete at 4 weeks was actually below the usual allowable
working stress of 1:2:4 concrete which is 500 lb. per sq. in. The
initial modulus of elasticity varied from 2,880,000 to 5,360,000
or approximately 100 per cent.

From these results, which it is believed are substantiated by
other field tests, it would seem that more attention must be
given to the mixing and handling of the concrete or it is not safe
to assume an ultimate compressive strength of 2000 lb. per sq.
in. for 1 : 2: 4 concrete at the age of 4 weeks.' This value is readily
obtained in the laboratory and by some contractors and there is
no reason why it should not be obtained by all. As a general
thing specifications are not sufficiently definite upon the method
of mixing and the consistency to be used and too much has been
left to the discretion of the engineer or foreman in charge of the
work.

Digitized by

Google

DISCUSSION

Mr.wason. Mr. L. C. Wason.^ — I would like to ask whether the labora-

tory or the field tests gave the higher results and whether they were
uniformly one way.

Mr. Wig. Mr. R. J. Wig. — The laboratory tests were always the higher

with one exception. There was one contractor who approached,
in fact met, the laboratory tests in most cases, but the laboratory
tests were always the higher and more uniform. There was less
difference between the government test pieces or beams.

Mr.wafon. Mr. Wason. — Speaking from memory, during the early

work of the Boston Transit Conmiission in subway construction,
tests pieces were taken from the actual work to the laboratory
and invariably gave higher results than the laboratory test pieces.

Mr. Humphrey. PRESIDENT HUMPHREY. — ^Thc objcct of the tcst was to deter-

mine if possible the relation between work done under labora-
tory and under field conditions. The three companies were
doing what might reasonably be called first-class concrete work.
The men, however, tried to do their best as they knew they were
going to the laboratory to do work under competition, although
an effort was made to eliminate this feeling.

The results show that no matter how conscientious a con-
tractor may be that unless he knows in some definite way what he
is doing, judgment is of no value in determining the quality of
concrete.

There is probably no phase of concrete work meriting so
much careful consideration at the present time as the very feature
brought out by the paper of Mr. Wig. It certainly is a subject
we have been hammering ever since this Association has been
in existence — better concrete. The assumption of 2000-lb.
concrete without knowledge whether it is obtainable, without
any attempt to determine whether it is obtained, cannot be too
strongly condemned. One of the essentials in concrete work,
which we cannot too strongly emphasize, is the development
of field inspection and tests of concrete.

(526)

Digitized by

Google

Discussion on Strength of Concrete. 527

Our Committee on Concrete Materials have not reported Mr. Humphrey,
recommendations this year, but we hope they will, because any
method is better than none. The best education any contractor
can have at the present time is to use some simple compression
machine, no matter how inexpensive as long as it gives approximate
results as far as the load is concerned, that will tell whether the
materials are giving a 1000- or 2000-lb. concrete, or as was shown
on the screen by Mr. Kinney, whether the strength is only a
little over 300 lb. per sq. in.

The influence of the percentage of water on the strength of
concrete is not fully appreciated. The concrete can be too wet as
well as too dry. It must be thoroughly mixed and water cannot
take the place of mixing. Use the minimum percentage of water
that you can to get a good consistency, a sticky consistency,
and mix a half a minute or a minute, if possible, in the mixer,
and the result will be a stronger and better concrete. The mixing
must coat the aggregate and rub it together which cannot be done
by merely adding water.

It is hoped that definite recommendations on standard field
methods will be made by the committee next year, not methods
that can be used in the laboratory, but methods that can be used
on the work. This is accomplished in Austria and other parts of
Europe through the control beam. It is very simple, but a rather
expensive field test for the average contractor.

Mr. Wig. — The concrete was all proportioned by volume Mr.wig.
and then weighed, that is, the volume measurement was trans-
posed into weight measurement. The sand was taken from a
bin containing a certain amount of moisture. The moisture was
determined before admission. The quantity of gravel and sand
to be used and the weight per cubic foot had been predetermined.
So these results of 1 part of cement, 2 of sand and 4 of gravel,
may vary slightly, depending upon the method of determining
the weight per cubic foot. The weight per cubic foot was the same
with either, practically, so all variables have been eliminated
except those to be compared: time, strength and workmanship.

President Humphrey. — There were two series of tests, Mr. Humphrey,
one on a 1 : 3 : 6 concrete and another on a 1:2:4. The three
companies had nothing to do with the preparation or the handling
of the materials. They simply fixed the consistency, which was

Digitized by

Google

528 Discussion on Strength op Concrete.

Mr. Htunphrey. Variable, by requesting the quantity of water. Some tests were
made by hand methods and some were made by machine mixing.
The average laboratory test on 1 : 3 : 6 concrete at 4 weeks was
about 1000 lb. per sq. in. The companies claimed this mixture
was not a practical mixture for reinforced concrete conditions and
the second series was made on 1 : 2 : 4 proportions under the same
conditions by the same companies. This concrete gave over
2000 lb. at 4 weeks.

Mr.chttbb. Mr. J. H. Chubb. — I would like to ask if there was any

record kept of the amount of water that the particular contractors
used, the consistency.* It seems to me that since the variation
was only in the long time tests, the result of those tests would
show that nearly all those variations were due to consistency.
In a long time test the variation was about half what it was in the
short time test. If they ran 60 days or 6 months it would probably
be much closer.

Mr. Humphrey. PRESIDENT HUMPHREY. — The amouut of Water used was

measured. While the contractor said how much water, we meas-
ured how much water he actually used, so that the tests when
they finally appear in the bulletin form of the United States
Bureau of Standards, will contain the actual amount of water
used in these various tests.

It is found that at the end of a year there is less variation than
at any earlier period. When it is considered, however, that
concrete structures are put into service in a very short time and
that the contractor wants to remove the forms some in 2 days and
most in 4 days, certainly a week, it must be known just how to
handle the concrete to obtain the strength desirable under such
conditions. The factor of safety generally increases with age as
the concrete grows harder, but the important thing to know is
how to make the concrete so as to get the maximum strength at
an early period.

Mr. Wig. Mr. Wig. — I think we have given a little too much attention

to the ultimate strength of concrete rather than to the yield point
and elastic properties — ^not so much the elastic properties as the
yield point. The yield point of concrete in these tests was at a
strength of 400 lb. in some cases at four weeks. The forms are
usually removed in one or two weeks. In such cases the yield
point would probably not be more than 150 or 200 lb. and it would

Digitized by

Google

Discussion on Strength of Concrete. 529

be a question whether the building would withstand its own load, Mr. wig.
its own weight. The advantage in this variation in the consist-
ency is in the variation of the yield point. When the concrete
yields it will ultimately fail. In putting the test piece in the test-
ing machine the load was applied so rapidly that the yield point
would be little affected and carried to failure. When the load is
left on a little beyond the yield point the concrete will ultimately
fail.

Usually the yield point will run around 1500 or 2000 lb.;
1000 lb. would be very ordinary, and 1000 lb. was obtained in
these tests.

Mr. Wm. M. Kinney. — I think that this brings us back again Mr. Kinney
to the question of aggregates. Some aggregates certainly do
affect the eariy strength* more than others, and that is some-
thing that must be studied; that is, these tests must be made
in order to find out the strength of the concrete in a certain length
of time. Referring to the paper by Mr. Cummings,* I believe
the use of steam resulted from the fact that in 14 to 16 days he
could not get sufficient strength in his concrete made from a
particular aggregate to drive the pile, which necessitated the use
of steam in order to cure it more quickly. A job usually can not
be delayed too long for foundations, and piles are the starting
of a foundation, of course.

*See p. 312.~£o.

Digitized by

Google

THE NECESSITY FOR FIELD TESTS OF CONCRETE.*
By Fritz von EMPERGER.f

One of the most essential precautions taken in every well
regulated manufacturing plant is in the control of raw material
as to its qualities in order to properly guard against poor material.
Failure to do this results in a reduction of the efficiency of the
finished work, causing either accident or reconstruction, which
sometimes has a far-reaching effect, especially in the case of con-
struction work where it often leads to the financial ruin of the
contractor even if there is no accident. It would seem natural
that in concrete construction this precaution should not be omitted.

At the present time, however, it is considered sufficient to
test the cement according to standard methods and to judge the
fine and coarse aggregates by visual inspection. The inadequacy
of this procedure has been demonstrated several times recently
through the fact that the concrete in parts of large buildings
resulted so poorly that they had to be rebuilt and in another
case the structure was accepted only at the combined risk of the
contractor and the owner because the only correct manner of
handling the matter was beyond their financial ability, that is
rebuilding. In all of these cases the raw materials, although not
of first class quality, were as good as those used in hundreds of
other cases where no trouble occurred. It was even proven
experimentally that the materials, claimed faulty by interested
parties, when mixed with first-chiss cement gave excellent results,
but when mixed with the cement used on the particular job, the
results were below the requirements. Similar results have pointed
to cement coming from certain mills. However, with first-class
aggregates this cement produced a concrete of suflicient strength.
In one case it was found that the cement had not been seasoned
sufficiently, which was immediately remedied. In this case the
control beam was used and it is mentioned to illustrate the
effectiveness of the check on the use of proper materials.

* Traoslated from the German, by the Secretary,
t Consulting Engineer, Vienna, Austria.

(530)

Digitized by

Google

Emperger on Field Tests of Concrete. 531

The difficulty with concrete control is the long time required
to obtain results. The speed with which modem concrete struc-
tures are erected makes each day of great value. Therefore,
the amoimt of time required to send the test pieces to a laboratory
should be eliminated because a quick control is possible only on
the job. The method given in the latter part of this paper makes
it possible to obtain results within from 8 to 10 days, which will
clearly indicate all poor concrete. It is hardly necessary to state
that in conctruction work it is only important to detect consider-
able variation of the concrete from the average quality.

In the above mentioned case the conditions were about as
follows: The aggregates were first class and the mixing and other
operations were properly done. The first 10-day test showed only
one half compressive strength and it seemed clear that the trouble
was most likely with the cement. In the meantime one story
had been built with such concrete. The test indicated that the
reduced strength of the concrete would still be sufficient and
precautions were taken only to secure the original quality of
concrete for further work. On completion of the structure the
story in question wa^ subjected to an exhaustive test. This
indicated clearly much lower results than in the case of other parts
of the structure; however the results were not low enough to
necessitate rebuilding so that the contractor came through with
a bad scare only. If the result of the first test had been much
lower, it would have been possible to tear down the one story in
time. This would have been a small expense compared with
rebuilding the whole structure.

In several cases of completed buildings the removal of forms
was delayed time and time again in the hope that the concrete
would harden sufficiently, which, however, it did not do. Such
are very exceptional cases and may occur in only one of a thousand
structures. One case, however, should be sufficient to warn the
contractor and to demonstrate the necessity of concrete control.
The extra expense is so little that it need hardly be considered in
determining the cost of a structure and offers insurance against
a disaster which may cause the ruin of the contractor's business.

In order to demonstrate the necessity of concrete control
reference to such exceptional cases is not required. There is to be
considered in first line the many small things affecting the manu-

Digitized by

Google

532 Emperger on Field Tests of Concrete.

facture of concrete which to this time have not received sufficient
attention. Their importance is much under-estimated and it is
consequently ignored that the combined action of these conditions
may have a very bad result. In the same way as it has been
customary to rely upon cement meeting standard specifications;
one has also relied on the richness of the mixture, although the
incorrectness of the assumption that the strength of concrete
is proportional to the amount of cement can easily be proven.
As soon as the mixture of sand and aggregate has reached the
point where its voids are fully filled by the cement no material
increase in the strength is to be expected. The extremely, rich
mixtures called for in many specifications are valueless and an
unnecessary waste of material. It is without doubt the business
of the contractor to determine the most suitable mixture of the
materials at his disposal, work which, in large cities, requires
occasional tests. It is clear, therefore, that the contractor must
keep control of the concrete if he desires to use the most economical
proportions, consistency and mixing of the materials available.
The owner, i. e., his engineer, or the city building inspector must
also have certain information along these lines so that no require-
ments will be specified which it will be impossible for the contractor
to meet. From several cases it can be proven that the required
compressive strength in official specifications for a certain pro-
portion of mixture cannot be obtained with the material available
or the reverse, that is, the exclusion of certain kinds of stone is
technically entirely unjustified. It is one of the worst assumptions
that one has intuition as to the strength of the material. This
jud^ent can only be obtained through the use for many years of
such tests as are described later.

Let it be assumed that the proper preparations in a special
case have been made and that the strength in compression at
8 and 28 days of the mixture of the selected proportions is known
under normal conditions. The one strength serves as the control
for the concrete; the other for the removal of forms. Now the
question is, which' extraordinary conditions on the work should
be considered. In the first case the most natural deviation from
the conditions originally stated would be the amount of water
added to the mixture and the procedure of mixing itself. The
compressive strengths determined on the test beams repeat them-

Digitized by

Google

Emperger on Field Tests of Concrete.

533

selves uniformly, which is not usually the case with concrete cubes.
This is then confirmation that no mistake has been made in the
manufacture of the concrete and that the completion of the work
represented by the test pieces can be anticipated without worry.
Even small deviations are a warning for more carefulness and
checking of the methods of making the concrete.

An entirely different picture is brought to us when there is
a question of the influence of abnormally high or low temperatures.
In the first case it is possible that the hardening of the cement will
be so rapid that concrete partially set is placed in the forms. In
the latter case the action of frost can withdraw the heat necessary
for the setting of the cement and destroy the concrete through the

■ ^"

FIG. 1. FIG. 2. — CROSS-SECTION OF CON-

TROL BEAM.

formation of ice crystals. These two possibilities cannot be repro-
duced in the laboratory. The contractor or the inspector must
be able to determine on the job just what is happening to the
concrete. The test pieces must be stored near the part of the
structure, the concrete in which is to be controlled and must
be of the identical concrete used in that part of the structure
which it is desired to check up. In order to make this possible
it is necessary to use a method which can be carried through
right on the job without much trouble and without a testing
machine.

In line with the thoughts given above, a short description
of the control beam used by the author covering all the points

Digitized by

Google

534

Emperger on Field Tests of Concrete.

noted is appended. The dimensions of the test beam are so selected
that the test piece will not be too heavy and can, therefore,
easily be handled, and so that no large load will be necessary
for the test. Further it has been borne in mind that the com-
pressive strength should be determinable by a simple calculation
from the breaking strength without complicated formulse. This
empirical rule changes according to the method of calculation,
which should naturally be the same as that used in calculating
the beams in the structure so that the stresses obtained can be
compared and the factor of safety directly determined.

In the following the usual method of calculation is used, which

7'47'-

FIG. 3. — ARRANGEMENT OF TESTING APPARATUS.

assumes the rectilinear relation of the stress and deformation and
the ratio of the coefficient of elasticity of the reinforcement and
concrete, n = 15, although this is a secondary detail and not directly
connected with the method of testing. The following equation
gives the strength of the concrete, the maximum bending moment
being M,

M = Cctn = Tim = <rcWc = -^x b m

Insofar as the notation is not shown in Fig. 1, h represents the
total height, h the width of the beam, and x the distance between
the neutral axis and the extreme fibre at the section in question.
According to Fig. 2 in our tests the values of h and h are each 3 in

Digitized by

Google

Emperger on Field Tests of Concrete.

535

Therefore, the effective cross-section /^ = 3x3 = 9 sq. in. and the
effective section of reinforcement F = J sq. in., and consequently
the percentage of reinforcement is 5.55.

The arrangement of the testing apparatus, Fig. 3, gives

M=:^X 29 = 14.5 P in. -Z6.

Calculation shows

x—2.1 in. and m^h =2.3 in,

FIG. 4. — METHOD OF MAKING CONTROL BEAMS AT VIENNA.

Therefore the extreme fibre stress

29P

2 X 3 X 2.1 X 2.3

= P

The value of P, the breaking load, is represented by the weight
of the carrjdng device and the bricks and f of the dead weight
of the beam (exactly 10.25/14.5). Having determined the load
P under which the beam has failed this figure represents the
value of (Tc in in.-lb. and is the desired measure of quality of the
concrete tested.

Further details of the method can be obtained from For-
scherhefi XIII, "Eine Guteprobe fur Beton" by Ing. G. Neumann,

Digitized by

Google

536

Emperger on Field Tests of Concrete.

as well as the author's paper presented in Berlin, 1911, to the higher
officials of the building departments. Session II, published by
W. Ernest und Sohn, Berlin. A description of an extensive
application of the method in the construction of K. K. Kriegs-
ministerialgebaude, Vienna, by Oberleutnant J. Kromus, appeared
in Beton und Eisen, 1911, Heft XIX and 1912, Heft I. Figs. 4 and
5 illustrate the method of making and testing the test beams on
this work.

In order to obtain correct results in such a method of test it
is only necessary to be careful in maintaining constant the values

FIG. 5. — CONTROL BEAM UNDER TE§T.

of Fi, the sectional area of reinforcement, in this case \ sq. in.,
and of A, the distance between top of beam and center of reinforce-
ment, in this case 3 in. Ever^i^hing else can be done so primitively
as would be the case without especial care on a job and it would
not affect the correctness of the results. According to experience
the results obtained are always more correct, reliable and more
quickly obtained than results from test cubes l)roken in a hydraulic
press. This method of testing represents an easy way for every
contractor to become acquainted with his materials and to deter-
mine whether he is obtaining the best results at the utmost
economy.

Digitized by

Google

DISCUSSION.

Mr. Arthur N. Talbot. — It would seem that if a satisfactory Mr. Taibot.
control test piece can be found, its use would be helpful to the
constructor; it would serve as a record of the construction. There
is a diificulty, however, in getting a beam or test piece which is
fully satisfactory. I have used a plain concrete beam of a little
larger crossnsection than mentioned here, 6 in. wide, 8 in. deep
and 36 in. in span. This of course will break in tension. It is
subject to the variations which we must expect in the tensile
strength of concrete and requires that more than one test piece
be made, as a single result might be abnormal and accidental.
Mr. Emperger has put in reinforcement in such a way as to make
practically a compression test, the beam being so long that there
is little danger of shear failure. Whether this form of test piece,
of the size shown, would be satisfactory in practice can be told
through experience with it. It would seem to me, looking at the
dimensions, that a test piece only 3 in. wide and 3 in. down to the
reinforcement would require such care in its construction that it
would not be suitable when the aggregate is of other than very
small size and that there would be objection to the size in practical
use.

President Humphrey. — It is a matter of interest to this Mr. Hamphrey.
Association to have this subject come up, as it is in line with the
endeavor of one of our committees to have recognized the neces-
sity of having some inexpensive field method by which the
quality of concrete can be obtained. I have seen this control
beam test made in Vienna and I feel that the care required to make
this test renders it impracticable. The section of the test piece
itself had a material bearing on the quality of the concrete and
the size of the aggregate also afi^ected in a large measure the
results obtained. I did not feel from the number of tests that I
saw made that there was any great concurrence in consecutive
results. It is a question whether or not the same degree of accuracy
cannot be obtained from a small cylinder or test piece made in a
simple compression machine. Certain it is that contractors

(537)

Digitized by

Google

538 Discussion on Field Tests of Concrete.

Mr. Humphrey, should try various methods of testing, with the hope that some
simple, inexpensive method for determining the quality of the
concrete may be devised. That to my mind is of the most vital
interest at the present time in reinforced concrete construction.

Mr. Brett. Mr. Allen Brett. — While as some claim the control beam

in the 3 by 4-in. section might not be accurate enough to determine
correctly the value of the concrete, yet, especially in running
concrete in cold weather, it seems to me that the control beam
would be sufficiently accurate to indicate when the concrete was
hard enough to pull the forms. I think this point should be more
emphasized.

Digitized by

Google

REPORT OF COMMITTEE ON TREATMENT OF
CONCRETE SURFACES.

The Committee has this year devoted itself to perfecting
specifications for stucco and to methods of testing compounds
for dampproofing of surfaces. Every member of the Association
who was known to have knowledge on these subjects was written
to, something over four hundred persons being addressed. There
was no real criticism from anyone^ and we therefore feel justified
in recommending this specification for adoption this year.

The specification as printed in Volume VII of the Proceed-
ings, 1911, should be changed as follows:

Page 587, near bottom, before paragraph "Sheathing Boards,"
insert:

"Frame. .The frame of building shall be so rigidly constructed
as to avoid cracking the stucco."

Page 589, near bottom, change title from "Brick or Tile" to
"Brick, Tile, or Cement Blocks."

Page 590, line 7, at end add "and thoroughly wetted."

Page 590, paragraph 6, change title to "Brick, Tile, or Cement
Block."

Page 590, near bottom, after title "Intermediate Coat," add
a sentence, "Intermediate coat may be omitted on brick, tile, or
cement blocks."

Page 591, near middle, change title "Fmal Coat" to "Finish
Coat."

Page 591, near bottom, change title "Finish" to "Surface
Finish."

Page 593, after paragraph 3 add:

Machine Stucco.

Stucco may be applied by a machine provided the results
obtained are equal to those produced by hand work.

The last report of. the Committee is amended by the
following:

Insert, page 579, near bottom, xmder V. Waterproofing.
(g) Method of testing dampproof compounds and coatings.

(539)

Digitized by

Google

540 Report on Treatment of Concrete Surfaces.

Introduction — ^As it is obviously not the province of this
Association to stoop to commercialism by publishing the names
and merits of any article of trade, your Coromittee below gives
the method by which anyone may make comparative tests of any
compound for himself.

In the tests of concrete waterproofers and dampproof con-
crete coatings, these materials are divided into the following
classes:

(1) Colorless Dampproof Coatings. — ^This class includes all
preparations intended for surface application for the purpose of
excluding dampness, which the makers claim will not change the
appearance of the surface to which they are applied.

(2) Black Waterproof Coatings, — ^This class includes all tar,
pitch, and asphalt preparations and other black waterproofing
paints. Such coatings should be able to withstand some pressure
head of water and are therefore termed waterproofers. They
would not be used where a decorative effect is required.

(3) Integral Waterproofers. — This class includes all materials
for waterproofing, whether pastes, powders or solutions, that are
incorporated into the mass of the concrete at the time of mixing.

(4) Colored and White Concrete Coalings. — ^This class includes
all coatings of a decorative nature intended for exterior use.
Many of the paints in this class are recommended by the manu-
facturer for their dampproofing qualities.

»
Method op Testing Dampproofing and Watbrprooping

Coatings.

This test is intended primarily for classes 1 and 2, although
many paints in class 4, when especially recommended by the
manufacturer as a dampproofer, are included.

The test piece is a 3 to 1 sand-cement mortar block 3J in.
square by 2 in. deep, with a depression in the top 2 1 in. in diameter
by about i in. deep made by inverting a telephone bell in the mold.
The mold is of wood, J in. oak, and is made in two right-angular
pieces which hook together at opposite corners, so that it is easily
removed by unhooking and drawing apart.

To obtain a porous block or one that will absorb 30 c. c.
of water in more than one and less than two minutes, the sand

Digitized by

Google

Report on Treatment of Concrete Surfaces. 641

and cement is gauged with from 9 to 10 per cent, of water by
weight (this per cent, being figured on the total weight of sand
plus cement) and tamped evenly into the mold. Some difficulty
may be found in making blocks of the proper porosity; that is,
just dense enough to absorb the 30 c. c. of water in not less than
one or more than two minutes; but after some experimenting
with different percentages of water and different degrees of
tamping, the proper result is obtained.

Denser blocks are made in the same manner, but the per
cent, of gauging water is higher and may run up as high as 20
per cent.

All blocks are kept in a damp closet for 24 hours and then
allowed to dry out in air for at least one week before being
used.

After the block has dried out, 30 c. c. of water are placed
in the depression and time required for the block to absorb this
amount of water is noted. The blocks, are divided into two
classes: those which will absorb 30 c. c. of water in more than
one minute and less then two, (these are called standard blocks),
and those which take less than one or more than two minutes to
absorb the 30 c. c. of water.

Those blocks which take up the water in less than one minute
are considered too porous for a proper test.

After this treatment the block is again thoroughly dried out.
The time required to absorb the 30 c. c. of water is recorded on
the block which is now ready for use.

The blocks which take up water in one or two minutes are
considered the standard block, but materials are tested on both
porous and dense blocks for comparison.

Each block is given a number for purpose of identification.
Porous or standard blocks are numbered consecutively from 1 to
500 and the denser ones from 500 up.

The waterproofing coating is applied only to the surface of
the depression and to the top surface of the block. Two full,
liberal and thoroughly brushed in coats are applied, the first one
being given at least two days to dry out before application of the
second.

Not less than one week after date of application of the second
coat, the block is ready for the first test.

Digitized by

Google

542 Report on Treatment of Concrete Surfaces.

Twenty-Four Hour Absorption Test.

The block is weighed dry. Thirty (30) grams of water are
placed in the depression. The block with the water in is allowed
to stand for 24 hours, with a watch glass over the depression, after
which time it is weighed again. The loss in weight is the evap-
oration. The water is then thrown out and the block wiped
dry and weighed again. The loss from the weight, including water,
is the absorption. The following is a tabulation of the dif-
erent weighings taken:

(1) Weight of block dry plus watch glass.

(2) Weight of block plus 30 grams of water plus watch
glass.

(3) Same as (2), but after standing for 24 hours.

(4) Weight of block plus watch glass, after water has been
thrown out and the block wiped dry.

Weighings:

(2) Minus (3) gives weight of water lost by evaporation.

(3) Minus (4) gives weight of water left in block after 24
hours.

The watch glass almost completely prevents evaporation
from the surface of the water. Before any appreciable amount
of the water in the depression can evaporate it must pass through
the waterproof coating and be evaporated from the surface of the
sides or bottom of the block.

The most eflSciently waterproofed blocks are those which
show the minimum amount of absorption and evaporation.

Test to Determine Time Required for Coated Block
to Absorb Thirty Cubic Centimeters op Water.

The 30 c. c. of water are placed in the depression in the block
and covered with a watch glass to prevent evaporation. A paper
is placed over the whole to exclude air currents. The block is
examined from time to time until all the water in the depres-
sion has disappeared. The time required for the block to absorb
the 30 c. c. of water is recorded. If all the water has not been
absorbed in 3 weeks time the approximate per cent, absorbed
in 3 weeks is recorded.

Digitized by

Google

Report on Treatment of Concrete Surfaces. 543

The length of time taken for the water to be absorbed by the
block is an indication of the efficiency of the waterproofing film
or coating.

The block is then exposed in the open air to find the effect
of weather exposure on the coating. It is brought in at intervals
of about 3, 6 and 12 months and the same tests as above described
repeated. The decrease, if any, in the efficiency of the coating is
noted after each repeated exposure.

Test of Integral Waterproofing.

For this test, the blocks are made up of a 1 cement, 2 sand
mixture plus the waterproofing compound, which may be paste,
powder or liquid. The waterproofing compound is incorporated
in the mix in the proper amount and according to the directions
as given by the manufacturer. The mixture is gauged, in all cases,
with the proper amount of water to give approximate maximum
density. The consistency resulting from this amount of water
gives a rather wet, quaking mortar, but one from which no free
water will rise to the surface on tamping.

This block is identical in size and shape to that described
above in the " Test of Waterproofing and Dampproofing Coatings."

It is kept in the damp closet for 24 hours after making and is
allowed to dry out in air for at least one week before being
tested.

The tests on these blocks are the 24-hour absorption and time
to absorb 30 c. c. of water tests and are identical with those
used for the waterproofing and dampproofing coating blocks,
described above in detail.

Description of Test of Concrete Coatings.

This class comprises, chiefly, coatings of a decorative nature,
intended for exterior use.

The paints iji this class are tested on the outer face of hollow
cubes, the faces of which are about 9 in. square, with walls from
1 in. to 1 J in. thick. These hollow cubes are made of a 1 cement,
2 J sand, 4 gravel mixture; and gauged to a rather wet consistency.
The block is made by packing this rather wet mixture into a 9 in. by
9 in. by 9 in. wooden form provided with an iron core about 12 in.

Digitized by

Google

544 Report on Treatment of Concrete Surfaces.

high and 6 in. square. The bottom of the block is made about
1} in. thick, the hollow iron core is then placed on this bottom
layer and centered up and the concrete packed in between the
core and the wooden form.

The core used has considerable taper so that it may be easily
drawn out after the concrete has begun to set up. After the block
has had 48 hours to harden up the wooden form is removed
and the faces are finished with a 1 cement, 2 sand mortar. In
finishing the faces the mortar is mixed rather wet and is merely
rubbed into the surface with a wooden float to give a smooth
surface. The block is ready for use not less than one week after
date of finishing up the sides.

Before applying any paints, the block is given a number,
markcki on the inside of one of the faces, and the faces are numbered
from 1 to 4, marked at the middle of the top edge. The date of
making the block is recorded.

A few minutes before applying a paint, the bottom third of
the face to be painted is thoroughly dampened by applying water
to it with a brush. Care is taken to wait until no visible water
is on the surface before the paint is applied.

In applying the first coat the paint is thoroughly brushed
in so that it will penetrate all small irregularities of the surface.
The second coat is applied not less than two days after date of
application of the first.

At the time of painting the fact is recorded that the bottom
third of the side was dampened. The following data are also
noted at this time:

Number of block.
Number of side.

Name of maker and name of paint (for both 1st and 2d coats).
Date of painting (for both 1st and 2d coats).
Per cent, of sediment in the paint can when it was opened
and consistency of this sediment.

Ease of application of paint. ,

Ease of mixing paint.
Hiding power of paint.

After the second coat has thoroughly dried, the general
appearance of the painted surface is recorded.

Digitized by

Google

Report on Treatment op Concrete Surfaces. 545

The points noted in regard to appearance are as follows: —

Dull or glossy.

Whether surface is smooth or granular.

Whether brush marks are visible or not.

Whether paint fills well or poorly, i, e., whether the irregulari-
ties of the concrete surface are well fitted up by the paint or
not.

After all four sides of a block have been painted (two coats)
and at least three days after the application of the final coat, the
painted surfaces are scrubbed lightly with a medium bristle
scrubbing brush with powdered soap and water. The effect of
this scrubbing is noted in detail, as to whether the coating is
softened, or whether the paint scrubs off, and if so, to what
extent.

After the scrubbing test the block is allowed to dry out and
after not less than one day from time of scrubbing it is filled with
water and kept filled for a period of 5 days. If the water leaks
through either the sides or bottom of the block to any appreciable
extent, the inside of the block is given a neat cement wash to
remedy this condition.

After the water has been in the block for a period of 5 days
it is emptied out and the condition of the coatings on the different
faces carefully examined and recorded.

Conditions resulting from the 5 days' water test to be noted
are as follows:

Softening of coatings.
Excrescence.
Discoloration of coating.
Cracking or blistering of coating.
Flaking off of coating.

After the block has thoroughly dried out the coatings of the
four faces are given a general comparative rating according to
their respective general appearance and condition.

The block is now ready for the weather exposure test, and for
this piupose it is exposed in the open air. The block is placed
with the bottom side up. Examinations of the blocks are made
at regular intervals of about 3, 6, and 12 months.

Digitized by

Google

546 Report on Treatment of Concrete Surfaces.

Conditions resulting from weather exposure to be noted are
as follows:

Chalking of coatings.
Checking and cracking of coating.
Scaling or flaking oflf of coating.
Fading or discoloring.

Page 574. After line 8 insert the following:

A Inethod of waterproofing has been submitted by one
member of the Association and is reported to have given satis-
faction. It is as follows:

For a flat surface lay a base of concrete while still wet and
plaster with i in. of neat cement troweled hard, then follow with
another layer of finishing concrete, the lower layer being at least
2 in. thick and the top layer 3 in.

On wall surfaces as soon as the forms are removed, thoroughly
wet surface, trowel on J in. of neat cement, and follow immediately
with 1 in. of 1 : 2 mortar before the neat cement has begun to
dry or appreciably set.

If the wall or floor treatment in this manner is large it must
be reinforced to prevent cracking.

Respectfully submitted,

L. C. Wason, Chairman,
Cloyd M. Chapman,
Alfred Hopkins,
Emile G. Perrot,
Henry H. Quimby.

Digitized by

Google

DISCUSSION.

The President. — There has been request for a discussion of The President,
the cement gun. * The principle of the machine is to deposit sand
and cement with compressed air, the manner of the deposition
being such as to obtain a density and compactness not possible
by hand methods. The sand and cement are first mixed dry.
It is one of the essentials that the mixture shall be dry, otherwise
it will cake in the hopper or so clog up the pipe that it will
not flow.

The character of the surface on which it is applied and its
preparation are of course the things which determine the per-
manency of the plaster applied. With this brief statement we
may begin the discussion by requesting Mr. Chapman to say a
word on the cement gun.

Mr. Cloyd M. Chapman. — The Westinghouse, Church Kerr Mr. chapman.
Company conducted some rather exhaustive tests last sum-
mer on the product of the cement gun, that is the qualities pos-
sessed by mortar applied by the gun were investigated. The
tests covered compressive strength, tensile strength, percentage
of voids, absorption through the surface, and adhesion to other
materials. In each of the tests the qualities were much superior
in the gun mixed mortars than in the best hand made mortars
made under similar conditions and of the same mixtures. The
method was to apply a 1-in. coating on the wall. The wall was
constructed with sheet metal, so hand plaster and gun plaster
could be applied to the wall an inch thick. A briquette which
would fit the regular tension machine was cut out with a die or
cutter. For compression tests a 2-in. thick coating of mortar was
cut into cubes with a die or cutter. The results of 28 different
tests were from 25 to 100 per cent better in the case of the gun
mixture than the hand mixture. The same applies to density
measured by absorption, surface absorption, absorption through
the surface film as well as total absorption on boiling 4 hours and
then cooling. It showed a very dense, high quality of mortar
and the uniformity was much greater than that of hand work.

• For additional information see Proceedings, Vol. VII, p. 5()4. — ^En.

(547)

Digitized by

Google

548 Discussion on Cement Gun Mortab.

Mr. ciupiiuui. The test pieces cheeked up closer and the leanness of the mix
that could be used was very much greater. Some beach sand
was obtained near the laboratory at Whitestone Landing, a very
high percentage of which would pass an 80-mesh screen; the
bulk of it, I think, would lie between 80 and 100 mesh. Propor-
tions of 1 : 6, 1 : 8 and 1 : 10 of that very fine sand gave pretty
solid mortar. No tests were made — and I want to emphasize
this — of the operation of the gun as a machine.

Mr. Brett. Mr. Allen Brett. — It is thought that the product of the

cement gun is automatically exact. It has been observed that
although placing into the receiving chamber all proportions of
materials, as 1 : 4 and 1 : 6 or 1 : 10, that the product on the
wall is approximately always the same. There always has to be
a certain amount of cement in the product on the wall and the
rest of the sand drops down. It is a very peculiar proposition.
The mortar on the wall always seems to have a certain amount of
cement, approximately 1 : 2^. Another point is the amount of
material required. The amount of material was estimated for a
certain job the same as for an ordinary stucco job. The sand
and cement were sent accordingly, but the material was all gone
before the job was half done. It takes about twice as much
material as in hand stucco or plaster. Of course the material is
there as the mortar is denser.

Mr. Chapman. Mr. Chapman. — One point was not brought out in regard to

the spatter and the failure of sand to adhere when lean mixtures
were used. We did not find that the spatter was a much greater
per cent of the total amount applied when lean mixtures were
used as against rich mixtures, but it was greater when coarse
sands were used. The trouble is caused by the larger grains
which have a momentum sufficient to rebound when the mixture
is not rich enough. Lean mixtures of coarse sand would rebouna
somewhat more than a rich mixture of coarse sand, and the rich
mixture would reboimd somewhat less than the lean mixture of
the same sand whether fine or coarse. A rich mortar forms a
putty-like surface to which the particles will adhere more easily
than in the case of a lean mixture.

The President. The PRESIDENT. — The first Stage in the application of the
mortar is a rebound of the sand until a sufficient coating of neat
cement is deposited on the surface in which the sand will bed.

Digitized by

Google

Discussion on Cement Gun Mortar. 549

Mr. Chapman. — ^That explains the strong adhesion of gun Mr. chApnum.
applied mortar. The process gives a coating that next to the
surface is quite rich.

Mr. Willis Whited. — There are a very large number of Mr. whited.
bridges in the State of Pennsylvania, the masonry of which is
built of inferior stone, which has suffered from the effects of the
weather; and other bodies of masonry, some of them quite exten-
sive, have suffered from other causes, such as bad foundations,
bad mortar, etc. To repair them with stone would spoil their
appearance. My idea was that the defective parts could be
replaced with concrete, and then the whole faced up with the
cement gun, so that they looked like concrete, and put in good
repair for very much less money than it would cost to tear the
whole wall down and rebuild it, thus making the whole wall
better, both in quality and appearance, than it was when first
built. Besides, much of this masonry was built in early times
when lime mortar was used exclusively. My understanding is
that mortar from a cement gun can be forced into open joints
quite a distance and do a good deal better work than any repoint-
ing could, and some of the masonry could be saved in this manner
that, otherwise, would have to be torn down and rebuilt. I
would like to find out whether my ideas are correct.

The President. — There is no doubt that the pressure under The President,
which the mortar is applied drives it into all the cracks, so as to
completely fill them. With the stone in proper condition the
mortar will stick to the surface, and with a first-class plasterer
who should always accompany the cement gun to finish the sur-
face as fast as the mortar is applied, a very smooth, pleasing sur-
face can be obtained. The gun can be applied a second time to
rough up, because the gun produces a pebbly surface, which pre-
vents hair cracks and gives a pleasing finish.

The great difficulty would probably be the removal of all
organic growth, otherwise the mortar would not have sufficient
bond and the frost would lift it off. The gun can be used as a
sand blast, to clean the surface first before applying the mortar.
I should think as a general proposition that the cost of reparing
those bridges would, even with the cement gun, be materially
less than tearing them down and replacing them entirely.

Digitized by

Google

550 Discussion on Cement Gun Mortar.

Mr. Heidenreich. Mr. E. Lee Heidenreich. — During the early days of the
cement gun I had occasion to investigate the same for the New
York Central Railroad at Grand Central Station in New York.
We borrowed a gun from the manufacturer and experimented to
ascertain how to cover structural steel with the gun. The great-
est trouble was in losing the material going past the edges; the
necessary thickness could not be obtained around the edges.
There was a great waste of material, although several kinds of
backing were placed behind the steel members; naturally con-
siderable waste would occur as the mortar would pass the side
of the steel member. Of course, at that time the gun was in a
very incomplete condition and, I understand, it has been improved
since. I would like to know if this waste of material in covering
structural steel with mortar for fireproofing purposes has been
diminished by some method.

The President. The PRESIDENT. — As far as the Chair knows the loss con-

tinues to be approximately about the same. It is interesting to
note that the material collected is clean sand free from cement,
which can be used again. The actual amount of material that
rebounds from the surface on which the mortar is being applied
remains the same.

Mr. Heidenreich. Mr. Heidenreich. — Taking for instance a wooden trough to

stop the stream of cement mortar, which passes the steel member
of course a certain amount of the charge of the gun passes down
along the trough to the bottom. Suppose I want to get to the
edge of a column or flange. More than one-half goes entirely by
the column, and that contains cement; it sets very rapidly, too.
The waste was fully one-half the total mortar material, including
what fell down between the flanges, along the edge of the columns
and material that passed by. This was one reason, at first, that
the cement gun was not considered economical for the purpose.
I do not know whether it was used since at the Grand Central
Station, but this was nearly two years ago.

The President. The PRESIDENT. — I am not quitc surc what the conclusions

of the New York Central Railroad oflficials were on that work,
but certainly the mortar was driven into all the little cracks and
crevices of the iron work and produced a tightness which could
not be obtained by any other method. The conditions in the
New York Central Station were unusual on account of the

Digitized by

Google

Discussion on Cement Gun Mortar. 551

extreme rapidity of corrosion of the steel. Some of the steel The President,
work, erected not over •two years, had rust on it to the extent of
^ to iV of an inch, and the use of some protective coating
to prevent further corrosion was a matter of great importance.
The matter of cost was not as important to the engineers as the
question of the efficiency of the coating.

Mr. Chapman. — ^As to the sand which rebounds from the Mr. chapmui.
cement gun, an analysis of two or three samples of the reboimd-
ing material showed about one part of cement to twenty parts
sand, although a 1 : 3 mixture was applied. Samples of the
material gathered up at the foot of the wall analyzed about 1 : 20.

Digitized by

Google

CEMENT COATINGS.
By F. J. Morse.*

The subject of cement coatings, or cement paints, as many
speak and think of them, is one which will receive a great deal more
of thought. As an example, in an advertisement after describing
a cement building from a fireproof standpoint, there is added,
"Painting not Necessary." This remark, as it stands and as it
is construed by the majority of users of cement, is correct and
really applies to oil painting as it is used for the protection of
lumber in building of frame construction. Good cement work
does not need any protection from rain or climatic conditions such
as soon destroy unpainted wood.

The difference between what is called painting or paint, or
cement paint and what cement coating really is and should be, is
that paint, t. e., an oil paint, when applied is only a film that lies
on the surface of the material applied to and has for a vehicle
linseed oil. Linseed oil is useless as a covering for cement or
lime because the alkali, acids and lime in cement, stone and brick
surfaces, and in the air, immediately attack the linseed oil. After
a year or two, at best, the life is gone and there is only a film
left, that under a microscope would be found porous, moisture
going through it. Water and moisture come in contact with the
lime underneath and the film cracks and peels, so of course paint-
ing is not necessary on cement, but absolutely useless. But
something is necessary, absolutely necessary for use on exterior
cement work. Large concrete structures are veneered with brick
or tile and lose their identity as cement buildings. Stucco homes
are built and after a short time the surfaces crack, sometimes large
patches fall away, they become very discolored; as the surface is
porous oxides from the air are carried in by the rain, dirt as well;
water goes through and corrodes the metal lath, and many times
entirely disintegrates it. On wood lath, moisture swells them,
nails rust, and the same trouble is had, cracking surfaces.

* Heath and Milligan Manufacturing Company, Chicago, III.

(552)

Digitized by

Google

Morse on Cement Coatings.

553

The first thing perhaps that comes to mind would be to
waterproof the dement so as to prevent the moistm-e and rain
from going in, discoloring the work, corroding the metal lath
and swelling of wood lath. This is the very thing cement coatings
are made for and they not only are waterproof but add with it
the decoration feature, in that most any color can be obtained
in several shades for selection from natural cement colors to green.

FIG. 1. — ROUGH CEMENT STUCCO RESIDENCE COATED WHITE.

Was originally non- waterproof, like adjoining house and now looks like a
white cement house; does not show painted effect.

yellow and reds. Many attractive color schemes can be worked
out, especially on stucco homes.

From the fact that cement coatings are applied with a brush,
like unto an oil paint on wood, many of them are called cement
paints and the majority of people among users and manufacturers
of cement, call it paint, so it is plain that either cement does need
painting, or cement coating must not be called paint. A real
cement coating must be so made and applied so that there is

Digitized by

Google

554 Morse on Cement Coatings.

penetration into pores of the surface it is applied to and not leave
a film, otherwise one reverts right back to an oil paint.

Cement coatings are in a measure going through an experi-
mental stage, by both manufacturer and user, for some of the
harder requirements made of them, and too, some cement sur-
faces will not take a cement coating. For instance, some cement
floors that are so hard and crystal-like that it is impossible to
scratch them, while others are soft and porous. Now on the soft,
porous floor a cement coating will harden and make waterproof
and prolong the wearing of the surface a very great deal. On
the hard surface that is not necessary, but the man who has this
hard floor wants some uniform or decorative color to it, as it
may be in a public corridor, dining room, etc. Unless the surface
is treated to make it porous so that cement coating could be made
to adhere, it would be useless to put it on because it would soon
wear oflF. This can be done by using a 10 per cent muriatic acid
solution as a wash to open up the surface so the coating can get
a chance to adhere and penetrate.

There have been cement floors so soft that the owners of the
building would not accept them and wanted the contractor to
take up an inch or two and lay a new floor (a very expensive job)
that have been made hard and dust proof by two coats of cement
coating. There has been one coat applied to an exposed brick
wall for 7 or 8 years, protecting the brick and mortar joint, still
in good condition. There have been swimming pools made water-
tight and enameled with cement coating. A finishing coat of
enamel in colors to match can be applied where enamel surface
is desired, and here is a point too of much interest. A cement
brick wall can be laid up and enameled with two coats of white
coating and one coat of white cement enamel at a cost of about
J6 per thousand brick, or $5 per square of 100 sq. ft., adding the
cost of the brick, say at $10 per thousand, there is a saving of
about $50 per thousand over the regular enameled brick, and
at the same time the joints are as well as brick. There have
been reservoirs holding 600,000 gallons of water, made water-
proof by two coats of cement coating, on the inside.

To cite an example of waterproofing done with cement coating,
one large corporation had built four big cement tanks, measuring
about 300 ft. around. They were square, 20 ft. deep, to store

Digitized by

Google

Morse on Cement Coatings.

555

soft coal in, filling the tanks with wat^r to keep the coal from air
slacking. The walls and floor after a time cracked and would
not hold the water. It was impossible to make a superficial
application on the inside surface, such as the membranous method
of waterproofing because the coal was taken out by a large grab
bucket or clam shell, and it would break the surface; it was impos-
sible to fill the cracks with cement, because to fill perpendicular

FIG. 2. — ROUGH CEMENT STUCCO OR PEBBLE WASH HOUSE, KENMORE AVENUE,

CHICAGO, ILL.

Cracks filled and entire surface covered with olive green cement coating.
Color is permanent and surface waterproof.

cracks with cement in a plaster form stiff enough to hold itself
did not have sufficient water for proper crystalization and bond-
ing, and would crumble and fall out, and grouted in, it would of
course run out. These tanks however were waterproofed with
cement coating, cracks filled with the coating in a heavy con-
sistency before being reduced and afterwards all surfaces covered
with the coating applied with a brush.

Digitized by

Google

556 Morse on Cbbibnt Coatings.

Another feature of cement coating is, that the cracks in a
stucco house can be filled in the same manner by the use of a
syringe with a spout about the size of the crack, the entire surface
afterwards being coated. Cement coatings are all classed and
described about alike, in that they are made for the waterproofing
and decoration of cement, stone and brick and stucco surfaces
and to make cement floors dustproof and stainless.

The word waterproofing covers a multitude of sins, and about
the first question is, will a cement coating waterproof a basement
wall on the inside to stop the water coming through the wall?
The answer is a very emphatic "no," as it is not made for that
purpose, no more than Portland cement is for filling teeth, or
mending chairs or china. Cement coating is to waterproof only
when the pressure is against it with the wall to back it up, for the
exterior walls above grade, against rains and snows and moisture
from the air and for decoration.

Some two years ago a rough pebble dash house was found
so discolored and cracked that the owner was unable to sell or
rent. The surface was so porous that the moisture had corroded
the metal lath and the iron oxide showed up on the surface.
This house has all been coated, cracks filled, and made pure white
and the surface made \yaterproof so it does not show stains or
discoloration.

In a recent popular article on concrete homes there is de-
scribed a house of cement plaster on Ijong Island Sound built by
one who spared no expense, and the surface is described as cracked
and apparently moth eaten. Another house near by is one that
had been erected by a small property owner and is of solid con-
crete, described as of a not very inviting color or texture. Such
examples can be seen in any of the suburban towns surrounding
the large cities, and is a further example of the necessity of more
educational work among the builders of cement buildings.

Cement in itself is naturally absorptive and in all building
work, where there is an exposed cement surface, not only from a
waterproofing standpoint but from a decorative point as well,
something is necessary. That something is a waterproof coating
such as a cement coating. Those who have studied the situation
carefully, noting the requirements, have been able to obtain
many excellent results.

Digitized by

Google

Morse on Cement Coatings.

557

Cement coating can be applied to cement surfaces, including
labor and material at about the following costs. On stucco or
pebble dash surfaces $2.50 per square for the first coat; about
$2 per square for the second coat. On sanded finished surfaces,
such as interior cement plastered walls or exterior, this work can
be done for approximately $2 per square for the first coat and
$1.50 per square for the second coat. On cement floors one

»i^-

■^^ 1 : ^S^^Tjm^^^^thw

■■yjyii

"itI

PIG. 3. — CEMENT PLASTER RESIDENCE, EVANSTON, ILL.

All cracks filled and entire surface made waterproof and uniform in color
with natural cement color coating.

coat can be applied for about $1.75 per square for the first coat
and $1.25 for the second coat. Two coats are recommended.
There are, however, many surfaces where one coat is sufficient,
as one coat of a real cement coating will waterproof and harden
the surface, but will naturally flat in differently in places according
to the porosity of the surface to which it is applied. The second
coat will bring the surface to a uniform color.

From the figures given above it can be readily seen that the

Digitized by

Google

558

Morse on Cement Coatings.

ordinary stucco house, which is. so universally liked and which
generally is of a size of approximately 35 to 40 squares, can be
coated with cement coating for about $200, a very small addi-
tional expense, as compared with the satisfying results that can
be obtained.

In the above article porosity is referred to as a mark of poor
concrete, the result of either poor cement or of wrong proportions

FIG. 4, — CEMENT STUCCO RESIDENCE AND CONCRETE SWIMMING POOL.

House covered with cement coating and swimming pool with white cement
coating and enamel.

in mixing. There is no question but that to a big extent this is
true, not so much as to the statement of poor cement for having
been identified personally with the cement business, I feel pretty
sure that all cements that are allowed to be shipped from the
mill have passed rigid inspection, and are good. Improper mixing
and lack of sufficient cement is, in a great measure the cause
of the necessity of waterproofing. A cement manufacturer
recently said that if a greater proportion of cement were used in

Digitized by

Google

Morse on Cement Coatings.

559

all work there would be less need of waterproofing. Another
thmg necessary is more careful supermtendence of work, as the
very first principle of waterproofing is density and this can only
be obtained by putting into the concrete sufficient quantity of
fine material to fill the voids in the aggregate. The better con-
crete and cement work is done, the less need there is for cement
coatings, from a waterproofing standpoint, but cement coatings

FIG. 5. — CONCRETE FIRE PRESSURE WATER TANK, 600,000 GALLONS CAPACTTY.

Made watertight with cement coating applied on the inside.

still add with the waterproofing feature the decorative feature, as
has been explained.

A beautiful white cement house can be had by applying pure
white cement coating with a brush to a cement house that has
been discolored and also it can be made soft shades of green or
buff, all with lime proof colors and permanent.

In applying cement coatings it is important that the surface
is free from dirt and grease and that it is porous, and that the

Digitized by

Google

560 Morse on Cement Coatings.

cement coating is used and applied in a thin enough consistency
to penetrate into the work. Cement coatings make cement work
more pleasing to look at, and at the same time protect it from
atmospheric conditions that soon plays havoc with it, especially
plaster and block work.

Digitized by

Google

I3ISCUSSION.

Mr. L. C. Wason. — Mr. Morse spoke of waterproofing tanks. Mr. Waaon.
I would like to know what pressure, how high a head of water it
withstood.

Mr. F. J. Morse. — ^This tank is one built for a fire pressure Mr. Morse,
tank for the International Harvester Company, holding about
6,000 gallons of water and 28 ft. in depth. They were not able to
keep a good pressure. That is the only test that we have had on
cement coatings, except in tanks used for storing coal.

Mr. C. W. Boynton. — It seems to me that some of the cases Mr. Boynton.
referred to by Mr. Morse should be explained more fully. I do
not believe that poor stucco work can be saved by painting over
the surface with a cement coating. Proper workmanship and
proper materials must be put into structures and one should be
careful not to be misled to believe that something beautiful can
be made out of stucco work that is fundamentally bad.

The commercial success of any cement coating is probably
dependent very largely upon poor workmanship in concrete con-
struction, or failure on the part of the builder to get the satisfactory
results and pleasing eflFects which can be obtained in many ways
without the use of such material. I do not want to be understood
as taking a position against all coatings, waterproofing materials,
etc., but I do feel that as cement users we should try to make the
best possible use of the material with which we are working. Lack
of knowledge as to how to handle concrete and mortar, lack of
interest in trying to learn, and lack of business energy and fore-
sight among cement users have resulted in something like sixty-
five waterproofing and coating compounds for cement work being
put on the market in the last four years. It takes money to keep
these businesses going. The cement user and his customers are
furnishing it. Just as soon as we learn to get anything like as
much out of the materials with which we are working, as there is
in them, this waste of money will stop. I believe, especially for
the coatings, there is a field, but it must be developed along the
line of imparting to concrete and mortar work some property or

(561)

Digitized by

Google

562 Discussion on Cement Coatings.

Mr. Boynton. element of beauty which is not attained by the efficient handling
of the materials. It is not sunmsing that the coating and water-
proofing business as a whole is looked upon very largely as a para-
site, for its development has depended to a great extent upon poor
work, and so long as poor work continues, the demand for these
things will continue; also the fact that poor work many times can
be temporarily covered up by the use of such materials tends to
encourage their use. This Association stands for quality of work
and that is what we all should strive for.

Regarding the treatment of floors, there are many poor
floors and we have not made much headway in solving the prob-
lem. In time we will understand the reason for dusting and vnll
be able to write a specification, which, if followed, will assure a
dustless floor. I base this statement on the one fact that dustless
floors are laid today. I am confident that when men who are
actually laying floors become interested and carefully note the
conditions which produce dusting, and those which prevail when
dusting does not result, we will be nearing the time when dusty
floors are a thing of the past. I am glad to know that until such a
time there are treatments which will keep down the dust, or
possibly make the floor dustless.

Mr. Morse told us that we could not use a grout to repair
cracks in a concrete wall because it would run through. Certainly
his coating must be near the consistency of a grout, yet it seems
to stay. I believe that if mortar had been mixed to the proper
consistency and applied in the proper manner it would have
stopped the cracks just as effectively.

Another thing Mr. Morse referred to was the making of denser
concrete by the use of fine aggregates. The government is about
to publish a bulletin (Technologic Paper 3, Bureau of Standards)
which shows that the densest concrete is obtained with aggregates
containing a very small amount of fine particles.

Doubtless Mr. Morse's coating and other coatings have their
place, which is in decorating a well built structure, and not in
covering up the defects in a poorly built one.
Mr. Morse. Mr. Morse. — I think I can answer in just a few words, as

stated in the paper, that if better concrete was made, or better
superintendence, there would be very little use for cement coatings.
Cement coating is not for good cement work; it is for cement

Digitized by

Google

Discussion on Cement Coatings. 563

work that is not made good. I think I tried to leave that impres- Mr.Mone.
sion with you.

Referring to the remarks about a part of a stucco house
falhng out. It was my understanding that stucco work was pebble
dash work, as we know, which contracts and expands. It is a
fact that the moisture beating through a stucco wall that is not
waterproof, continually keeps up a corrosion of the metal lath,
and the same entirely disintegrates. It is at such places where the
wall falls out, and I do claim that a waterproof cement coating
applied to that wall would prevent expansion and contraction to a
large extent and the corroding of the metal lath.

Another question raised was filling the cracks with cement.
A perpendicular crack with no bottom, a crack in any concrete
wall, for example, or stucco wall, it is impossible to grout with
cement. The cement will keep running out. Using a plastic
cement stiff enough to hold, in setting the two sides of the cement
will withdraw or dehydrate, taking the water out of this cement
and not enough water remains for proper crystallization. That
has been my personal experience. The cement coating is of a very
heavy consistency, and put into a perpenilicular crack will harden
there perhaps better than cement. In this particular work cement
had been tried and failed, and the grouting was a success with
cement coating.

The question of taking care of cement floors is one that has
caused a great deal of trouble. With a hard floor that is very
crystalline, you refuse to put anything on it, but on a floor that
dusts up it must be covered.

The placing a cement floor that does not dust is to my mind —
we are all looking for information in this work — almost impossible,
especially for a man to guarantee or assure a floor will not dust.
There is a psychological moment to trowel a cement floor which
can only be determined in the laboratory, and that is immediately
after the initial set has taken place. The action of troweling a
cement floor forms a suction and that draws the fine particles to
the surface. Now if the troweling is done after the initial set has
taken place it breaks up the crystallization being formed and
dusting of the surface is probable.

As to the outside cement surface in reinforced concrete build-
ings, I have seen some work that has been rubbed down with

Digitized by

Google

664 Discussion on Cement Coatings.

Mr. MorM. carbonindum stone in panel effects that are beautiful, and it would
be a crime to apply a cement coating to them.

Mr. Boynton. Mr. C. W. Boynton. — I think Mr. Morse covered the whole

thing when he said cement coatings are not necessary in good
work. I most heartily agree with him. This Association, if it
stands for anything, stands for good work. It is exactly what we
are aiming at and what we are coming to. However, I am glad
that Mr. Morse and his company stand ready to throw themselves
into the breach and help reclaim the poor jobs.

The Preddant. The PRESIDENT. — There is probably no more important field
than the subject of a color of some kind that can be applied to a
concrete surface so as to be permanent. Most colors are more
or less injured by the cement itself. They leach out so that after
a very short time the color is gone. The Germans have a way of
treating the concrete with an acid wash, which neutralizes the
surface, and on this surface durable color can be placed. The
dampness that is often found on the interior of concrete walls is
in large measure overcome, and I have never seen damp walls
on the interior of foreign buildings. They are generally insu-
lated in some way with air spaces, so that the dampness with
which we are troubled in a great many cases is lacking in those
structures. Certain it is that the treatment of the interior so as
to produce pleasing color efifects is one of the most important
steps that must be considered.

It is perhaps easier to tint the interior of a structure than it
is to tint the exterior, and some experiments which have been
made by the Committee on Treatments of Concrete Surfaces
have shown that the various pigments, when they are put out in
the weather, soon lose their color and coherence and practically
become worthless. So that the whole subject of the treatment
of the material, the conditions to be observed in applying color to
concrete, is very important and very difficult.

The tendency to use selected aggregates as a color and to
secure color efifects through the material itself, I think, is prob-
ably the most hopeful field. The use of white cement and an
aggregate which possesses color value will give tints that cannot
be duplicated with ordinary pigments. Perhaps the best example
of color effect is to be found in the Connecticut Avenue Bridge in
Washington, D. C, where the Potomac River gravel concrete of

Digitized by

Google

Discussion on Cement Coatings. 565

buff color is in contrast with the concrete which is made out of The Preddent.
the gneiss rock of bluish color found on the site. The spandrel
walls, etc., in buflf are in contrast to the archrings and piers in a
light blue color; and the color is distinctly visible for a consider-
able distance. These colors of course are permanent. The most
permanent colors, as far as the exterior exposure is concerned, are
due to an aggregate which is of itself permanent. That is, red
can be obtained by granites and some forms of hard sandstone,
which give us a permanent color effect, blue from traps and lime-
stones, yellow from sandstones and gravels.

The great difficulty in securing color effects in concrete and
mortar is that the cement itself has a muddy color, and perhaps
the sand is also of a dirty color, resulting in a mottled appear-
ance, because it is difficult to add enough coloring material to
overcome the muddiness produced by the sand and cement.
The use of white cement and white or light sand makes it possible
to add a very small percentage of color or to use an aggregate
with color and secure the desired color effect. The study of this
subject and the grading of the material and selecting it with the
view to securing color effect is one of the developments which
will characterize our future work.

Digitized by

Google

REVIEW OF THE PRESENT STATUS OF
IRON ORE CEMENT.

By p. H. Bates.*

It is generally understood that there are three essential con-
stituents in Portland cement, namely: lime, alumina, and silica.
It has been shown experimentally that the lime can be entirely
replaced by strontium or barium, the alumina in part by iron or
chromium oxide, and hydraulic materials of considerable value be
obtained; but none of these have been suggested as being of
commercial value with the exception of iron oxide.

There exist very extensive deposits of low grade ores in which
the percentages of iron are too low to admit of their economical
reduction in the blast furnaces. There are also numerous deposits
of high iron oxide clays in which the high percentage of this con-
stituent reduces the value of the clay. Both of these have appealed
to the cement manufacturer but to a slight degree, owing largely
to abundant supplies of clays of low iron content, from which it is
comparatively easy to produce cement of normal composition.
The manufacture of a cement of high iron content presents some
difficulties in view of the low fusion point of the raw material,
there being very readily produced the so-called — from the point of
view of the manufacturer — overbumed cement.

There had been noticed in several localities abroad, the failure
of concrete made from cement of normal composition when placed
in sea water. Whether this was due to faulty cement is an unset- J

tied question. But various authorities on casting about for an |

explanation of these failures quite unanimously came to the con-
clusion that the trouble was due to the alumina in the cement.
This constituent combines with the sulphuric anhydride of the sea
water and the lime, freed in the process of the setting, to form a
compound of indefinite composition. This latter crystallizes with
a large water content, which pushes apart the cement particles
and ultimately destroys the structure.

Necessarily the first remedy which suggested itself was the

♦Chemist, Bureau of Staodards, Pittsburgh, Pa.

(566) I

I
I
I

Digitized by VjOOQIC j

Bates on Iron Ore Cement. 567

replacement of the alumina with some material which would not
have this property. Michealis, reviving the original suggestion of
Malaguiti and Durocher, called particular attention to the re-
placement of the alumina with iron oxide. Attracted by Michaelis'
arguments, the Krupps in 1901 took patents covering the manu-
facture of such a material.

There is no doubt but that the alumina present in cement
will form a compound with the sulphuric anhydride radicle of
soluble sulphates and with lime hydrate. But the exact nature of
this compound is still much in doubt; just as doubtful is the ques-
tion whether the compound can result from the setting of all the
alumina constituents which may be present in cement. It is very
generally agreed, however, that the iron compounds which are
present in a cement cannot form such a decomposition product, but
whether the iron is combined in the cement in a manner
similar to the alumina is a very doubtful matter. The latter
problem is complicated by the impossibility of securing materials
for investigation which are so sufficiently free of alumina as not
to be decidedly influenced by the presence of small quantities
of the latter. Thus the Geophysical Laboratory has shown that
0.5 per cent, of alumina present in a silica lime mix of such a com-
position as to give tri-calcic siUcate, will give the maximum yield of
this compound. It is quite likely that those who claim the iron
compounds in cement have hydraulic properties, have been mis-
led by the influence exerted by the very small quantities of alu-
mina which have been present in these materials. Without doubt
the iron compounds lack hydrauUc properties entirely, and the pos-
sibility of making a cement entirely lacking alumina will never be
attained. The iron ore cements must have alumina present or else
they are of no value, and in the finished cement the iron oxide is of
no other value than any other material which might be present
which would reduce the alumina content. This statement does
not deny the value of the iron oxide as a flux in the manufacture.

It is therefore not surprising that the manufacture of this class
of cement has not grown, even in Germany, as was hoped for. Its
use, even in sea water, has not kept pace with the growing use of
the "Eisenportlandzemenf (or "Slag-Portland'^ cement) or the
use of natural slag (Puzzuolana) in connection with PortUnd
cemejit. The theory of the use of the latter is based on the fact

Digitized by

Google

568 Bates on Iron Ore Cement.

that the lime which is set free in the setting of the cement, will
combine with the silica of the slag or puzzuolana, thereby rendering
less lime available for combining with the sulphuric anhydride
radicle of the soluble sulphates, and consequently less possibility
of the forming of the injurious sulpho-aluminate of lime.

Notwithstanding the discussion and notices which this class
of cements has received in the technical journals, there is little
authentic data available showing its relative superiority over
cement of normal composition. Thus, Michaelis, Jr., quotes from
the ZentrcdblaU der Bauverwaltung in the Cement and Engineering
News of March, 1911, the following compressive strengths of
an iron oxide and Portland Cement in the form of 16-inch cubes:

I I Compreaoiva

Paris of Cement. i Parts of Sand and Cniflhed Stone. ' strength.

I . (lb. persq.in.).

Iron ore 2

" " 2

it tt 4

Portland 2

3 ' 6776

3 6476

6 3120

6 , 2910

3 4690

2:3 6146

4 6 I 2416

4 ' 6 I 3270

The results were obtained by the Department of Public Works
Berlin at Swinemiinde and represent the strength after three years
immersion in sea water. While Michaelis concluded from these
results that the iron ore cements are the stronger after such treat-
ment, these conclusions are hardly justified, since the above may
represent the relative strength before treatment. However, the
description of test pieces, both at this point and at other localities
along the German sea coast, undoubtedly shows the superiority of
low alumina cements.

Quite a number of tests of iron ore cements are given in the
Toninduatrie-Zeitung (1906, page 1968). The comparison of them
with normal composition cements was carried out entirely by the
use of small prismatic test pieces which were placed in the sea water
and their appearance noted. Another series was conducted in
which different amounts of plaster were added to the cements from
which the prisms were made. In every case the resistance of the

Digitized by

Google

Bates on Iron Ore Cement. 569

iron ore cements to the sea water was greater than that of the Port-
land cements. Unfortunately, comparison by means of tensile or
compressive strengths was not made. E. Maynard in the same
joiunal (1910, page 651) conducted somewhat similar tests; he
however showed the slight change in chemical composition which
the iron ore cement underwent by such treatment, but does not
give results for Portland cement.

In France there has been little investigative work carried on;
however, Candlot* gives some interesting results obtained with five
high iron oxide cements manufactured by himself. The partial
chemical analyses and the strength of the neat and 1 to 3 sand bri-
quettes at the end of the 7 and 28 day and 5 year periods, after im-
mersion in water, are:

No. 1. No. 2. No. 3. No. 4. No. 6.

SiO, 19.0 21.0 26.0 24.6 26.0

AljOi 6.8 5.3 5.3 6.3 6.4

Fe^, 7.2 6.2 8.8 7.2 6.6

CaO 66.6 66.0 69.0 61.8 62.6

98.5 97.5 99.0 98.8 98.5

Tensile strength, pounds per square inch.

No. 1. No. 2. No. 3. No. 4. No. 6.

Neat Briquettes. . 7 days 951 845 142 220 426

28 " .... 1,036 - 913 213 668 724
6 years.... 653 824 894 769 717

1:3 Sand 7 days.... 356 511 92 234 356

28 " .... 398 525 199 398 497
5 years.... 678 710 646 724 667 (strength at

end of 3 yrs.)

The author calls attention to the high lime of number 1 and
mentions that the pats in steam disintegrated, although the bri-
quettes showed no such signs; number 2 has a silica and lime
content corresponding to a cement of normal composition; both of
them show the retrogression in strength of the neat cement bri-
quettes, after the 28-day period, so characteristic of cements of
normal alumina content. The three other cements are really high
silica cements and show the characteristics of this class very

* h9 Ciment, May, 1011. page 32.

Digitized by

Google

570 Bates on Iron Ore Cement.

strikingly in the tensile strength — namely the very low initial
strength followed by satisfactory gains.

The effect of increasing the silica content with corresponding
decrease of the lime content is shown by numbers 1 and 3, and 2
and 5. Numbers 3, 4 and 5, which show decreasing amounts of iron
oxide, show increasing strength at the early periods. However,
these three cements show too much variation of the other constit-
uents to attribute this gain to decreasing iron oxide.

In this same paper Candlot gives the partial chemical analyses
and tensile strength, at early periods, of some silicious cements
which he manufactured.

No. 1. No. 2. No. 3. No. 4.

SiO, 26.0 25.0 27.0 26.0

AltOa '. 2.6 1.7 3.4 2.0

CaO 67.0 66.0 65.0 67.0

95.5 92.7 95.7 95.0
Undetermined 4.5 7.3 4.3 6.0

Tensile strength, pounds per square inch.

No. 1. No. 2. No. 3. No. 4.

Neat Briquettes.. 7 days 899 894 692 777

28" 965 909 885 866

1:3 Sand 7 days 601 398 426 440

Briquettes 28 " 667 635 586 535

These strengths hardly correspond to those of such high silica
low alumina cements. In view of the large percentage of unde-
termined shown in the analyses, it would appear as if there were
present some other constituent, not determined, which had caused
these abnormally high strengths. In the discussion following the
reading of the paper, Candlot acknowledged the use of a flux —
"generally iron oxide" — in their preparation. We have here con-
sequently a group of high silica, relatively high irqn ore cements,
which are of much interest.

No information is given in regard to the superiority of these
several cements of Candlot's over normal Portland cements, when
immersed in sea water. They do seem to show the high strength
which it is possible to reach with high silica cements and usually
the steady gains in strength, Thege high silicious cementa

Digitized by

Google

Bates on Iron Ore Cement. 571

should be receiving more attention by the investigator, the manu-
facturer and the consumer. That Portland cement is overlimed
and has principally but the ease of manufacture and ease of han-
dling by the consmner to recommend it, is a growing conviction.
The latter statement is substantiated by the increasing use of
trass, tufa, etc., which use would be impossible were not normal
Portland cement overlimed.

Until within a very recent period there has been no attempt in
this country to manufacture a high iron Portland cement and place
it on the market as such. There have been on the market for a
number of years, however, two cements, manufactured here, which
are of exceptional high iron oxide content, although this fact has not
been used by their manufacturers for recommending them for other
than the ordinary purposes. In the two cements referred to the
percentage of iron oxide is little less than the alumina.

There has been just as much inattention in this country
devoted to the investigating of this class of cement as there has been
to its manufacture. About a year* ago there was conducted a series
of tests at the University of Illinois in which it was desired to
show the superiority of this class over that of normal composition
cements for sea water use. The high iron cement used was manu-
factured in the Ceramic Department of the University, and was
compared with a normal Portland cement secured on the market.
According to the tests reported, which consisted of making pats
and briquettes of the two cements and treating with sea water
under pressure and observing the disintegration and loss of strength,
the former was superior to the latter. But other burnings of high
iron oxide cement made here gave the opposite results; the author,
however, explains this by improper composition of his burnings,
so that on the whole his results are not very conclusive. He
however did find that alumina was absolutely essential; a burning
in which it was entirely replaced by iron oxide did not give a
material which could be called a hydraulic cement.

The only other extensive series of tests made in this country
are those which were carried out at the Atlantic City Laboratory
of the Technologic Branch of the U. S. Geological Survey, now a
part of the Bureau of Standards.

In Table I are given some of the results obtained, using not

• See p. 597. Ed.

Digitized by

Google

572

Bates on Iron Ore Cement.

Table I. — ^Tensile Strength in Pounds per Square Inch op Neat

AND 1 : 3 Sand Briquettes of Various Cebients When Immersed

IN Fresh and Sea Water.

Slag Cbme>jt (E).

Natural Cement (I).

Neat Briquettes.

1 . .

Fresh , Sea
Water. | Water.

4 weeks ....
13 " . . . .
26 " . . . .
62 " . . . .

..' 549 1 686
658 730

. . 1 634 662
422 432

Neat Briquettes.

Fresh
Water.

282
368
378
366

Sea
Water.

302
368
236
283

Fresh
Water.

Sea
Water.

116
196
182
260

Ill

166

208

Mixture, Natural and Portland Cements (J). Portland Cement (K).

1 Neat Briquettes.

Sand Briquettes.

Neat Briquettes.

Fresh Sea
Water. Water. 1

Fresh
Water.

247
201
267
218

Sea
Water.

Fresh
Water.

Sea
Water.

4 weeks ....
13 " . . . .
26 " . . . .
62 " . . . .

221 1 258
216 311
204 286

. . 1 260 244

190
256
318
276

866
910
830
463

1100
927
900
923

Portland Cement (L). Portland Cement (M).

4 weeks
13 "
26 "
62 "

Fresh
Water.

595
557
550
407

quettes.

Neat Briquettes.

Sea
Water.

Fresh
Water.

Sea

Water.

623
607
673
233

748
630
610
607

837
799
670
277

Slag Cement (F).

Neat Briquettes.

Sand Br

Fresh
Water.

iquettes.

Fresh
Water.

Sea
Water.

468
512
348
210

Sea
Water.

4 ^fkeks

416 1

210
306
246
336

180

13 "

511

241

26 "

590 '

268

62 "

. . . . ' 636 1

280

Digitized by

Google

Bates on Iron Ore Cement.

573

Special Low Alumina Portland
German Iron Ore Cement (A). Cement (B).

Neat Briquettes

Sand Briquettes

Neat Briquettes

Fresh Sea
Water. Water.

Sand Briquettes

Fresh
Water.

535
714
726
764

Sea
Water.

702
746
802
656

Fresh
Water.

120
221
234
256

Sea
Water.

134
166
236
204

Fresh
Water.

Sea
Water.

4 weeks

1
764 714
803 554
800 622

768 1 748

242
311
418
370

252

13 "

355

26 "

320

52 "

310

Portland (C).

Neat Briquettes

Typical Portland No. 47 (G).

4 weeks
13 "
26 "
62 "

Fresh

Sea

Water.

672

748

744

808

708

768

636

764

Sand Briquettes

Fresh

1
Sea Fresh

Sea

Fresh

Sea

Water.

Water. 1 Water.

Water.

604

821

268

206

1 636

136

300

238

, 658

304

258

674

• *

350

205

Typical Portland Cement No. 47 (G).

Specimens stored in damp ] Specimens stored in damp
|Clo8et*for 24 hours; then for closet for 24 hours: then for
2 days in fresh water. 6 days in fresh water.

Neat Briquettes.

Fresh
Water.

Sea
Water.

1 Fresh
1 Water.

Sea
Water

4 weeks

672
765
713
622

959
904
852
738

700
723
734
660

926

13 "

840

26 "

940

52 "

598

White Portland (H).

French Portland
Cement (D).

Sand Briquettes.

Neat Briquettes.

Sea
Water.

Sea

Fresh

Water.

164

539

200

162

702

204

776

511

656
706

* A second lot of briquettes gave substantially the same results.
** Broke in handling, before placed in testing machines.

Digitized by

Google

574 Bates on Iron Ore Cement.

only this class of cements but in addition for the sake of compari-
son, Slag, Natural, White, Slag-Portland, and normal composi-
tion cements. Among the latter is a well-known French Portland
cement which while of normal composition is highly recommended
and rather extensively used for sea water concrete. Its distinctive
characteristics are the low sulphuric anhydride content and its
decided coarseness, 21.3 per cent, being retained on the 100 sieve
and 38.8 per cent, on the 200. Unless otherwise noted, all the
briquettes were stored for 24 hours in the damp closet before
immersion in the waters. In Table II is given the chemical
analyses of all cements, the tensile strengths of which are reported
in Table I.

An examination of Table I shows that the neat briquettes
made of Portland cement of normal composition retrograde
considerably in strength after the 13 weeks period. In this
class have not been placed the cements K and C." since both
of them show rather high iron oxide content, although they
could hardly be called iron oxide cements. Cements L and M
were simply chosen at random from standard Portland cements,
while the typical Portland is a mixture of equal parts of a large
number of standard Portland cements. The efifect of aging, even
for a short period in fresh water before subjecting the cement to
the action of sea water, is very closely shown, and suggests a
practice which should be observed, when at all possible, in all
cement used in sea water. The rather remarkable behavior of
the White Portland, in view of the theory under which high iron
oxide cements are used, is striking and unexplainable. The data
on E has been given largely because this cement, if we are rightly
informed, belongs to that class of cements called by the Germans
"Eisenportland" cements, which are mixtures of slag with Port-
land cement, though contrary to the German practice more slag
has been used in its manufacture than Portland cement. This
cement, while it shows retrogression, does not show the decrease
in strength which the true slag cement does.

The sand briquettes of all the classes compare rather favor-
ably, and it would be difficult to form an opinion of the relative
value of any class from the behavior of the mortars alone. With
the exception of the Cement J, all the sand briquettes show less |

strength in sea water than in fresh water at the end of a year; '

I
I
I
I

Digitized by

Google

Bates on Iron Ore Cement.

5^5

CO 05 CO

s s

1^ t^

CO »o

»0 lO

« «5 g

C^ rH

•

2 S S

g s?

s; §

S 8

8 :

« t- g

CO i-H

1-H

S§

R f: ^

g 5

s »

S? g

g : :

1-H

n to m

rH rH

^ 8 g

S J2

s s

?5 !::

rH
rH

c « gj

c3 *^

^ s

i-H

S 5? So

: §

gg : :

rH to 1-t
CO

1-H

»-H

18 S «

2 S

5 3

!§ 3

S : :

«> g

»H 1-H

2 ^

8 2 S3

S§ 2

g :

1-4

M X C^

CO fi

• •

S i 2

t 2

§ S

§5 : S

-" 3 §

<N rH

CO • <N

f: 8 fe

1-i

• o

: S

S 2

Tji

• M

1-H

§5

t>.

1-H

£2

Csl

1-H

■^

»H

Tji

rj*

1-H

8 :

1-H

r-i

S 3

1-H

»-<

»o

■1

: ^

: <2

D n

• 1

• S

■!^

Iz;

^ C

Digitized by

Google

576 Bates on Iron Ore Cement.

those of cements I, F and White at 1 year do not show a retro-
gression since the previous period of testing.

The relative strength in compression of concrete made from
German iron ore cement and normal Portland cement is shown
in Table III. All specimens were stored for 3 weeks in a damp
closet before immersion in the water, and while the concrete

Table III. — Comparison of Comprbssive Strengths of Concrete

MADE FROM AN IrON OrE AND A PORTLAND CeMENT.

German Iron Orb Cement (A).

Compressive strength in pounds per square inch (1 part cement, 2 parts
sand, 4 parts trap rock).

Stored 8 weeks in damp closet before immersion in —

Fresh Water. Sea Water.

13 weeks, 3743 3477

26 " specimen 3987 A B

52 " B B

A. Two of three specimens did not break under a load of 4237 pounds,
capacity of the testing machine.

B. All specimens did not break under a load of 4237 pounds, capacity
of testing machine.

Typical Portland Cement No. 47 (G).

Compressive strength in pounds per square inch (1 part cement, 2 parts

sand, 4 parts trap rock).

Stored 8 weeks in damp closet before immersion in —

Fresh Water. Sea Water.

13 weeks, 3190 3877

26 " 3457 3979

52 " 3389 4060

from the former cement is stronger than the latter, yet it is not
very decidedly marked. At all periods the Portland shows
greater strength in sea water than in fresh water, whereas, at the
early period the Iron Ore shows a less and apparently at the later
periods about equal strengths in the two waters.

These abstracts from the data of the Atlantic City Laboratory
show in a general way the relation of the strengths of the Iron Ore
to various other cements. Again, it cannot be positively stated

Digitized by

Google

Bates on Iron Ore Cement. 577

that the former show much superiority over the latter classes of
cements. The information to be obtained in foreign publications
is usually of a descriptive character and is concerned but little
with numerical data, which of necessity are the only positive and
reliable data. In this respect the information and data are much
different from that obtainable concerning the "Eisenportland"
cements, and if one is to judge of the use of the latter from the
numerous investigations which have been and are still being
carried on in regard to its general use and use in sea water concrete,
it must be a very important and valuable product, far more so
than the "Erzzement" or Iron Ore cement.

Digitized by

Google

MARINE OR IRON ORE CEMENTS.
By Herman E. Brown.*

As soon as Portland cement became an accepted building
material, engineers and chemists commenced investigations as
to its behavior under the varying conditions to which it was
exposed. The object of these investigations was two-fold — ^to
assure themselves of the permanency of the structure already
erected, and to find the outer limits of the field in which it could
be used.

As early as 1854 Malaguti and Durocher (Campt Rendvs,
Vol. 39, p. 183) were investigating the cause of disintegration of
concrete masses in sea water. As a result of such investigations
on commercial hydraulic cements, and on synthetic cements
manufactured by them, they state: "The inertness of the oxide
of iron in the hydraulic materials appears demonstrated by these
synthetic tests. We are forced to conclude that the presence of
this oxide is able to enhance the stability of cement mortars in
sea water."

Malaguti and Durocher had been inspired to make these
investigations on account of the discoveries of Vicat in 1840,
who had found that a certain hydraulic cement, after six months
in sea water, had been greatly damaged, an analysis of the cement
showing that the magnesium had been largely increased and the
calcium oxide decreased by about the same amoimt.

In 1882 Landrin makes the following statement: "The
calcium aluminates in cement are extremely harmful for perma-
nency of mortars in sea water, because they dissolve easily."
(Thon IndiLstrie Zeitungy 1882, p. 177.) Other prominent investi-
gators now began to take a lively interest in the search for the
exact cause of the occasional sickness of concrete structures in
sea water. Le Chatelier and his co-workers commenced a series
of painstaking experiments. In 1890 Le Chatelier writes: "The
double sulphate of calcium and aluminum plays an important

* Chief Engineer, American Cement Engineering Company, New York, N.Y.

(578)

Digitized by

Google

Brown on Marine or Iron Ore Cements. 579

role in the destruction of concrete in sea water." Candlot in
1892 discovered the double salt of calcium and alumina. In
1899 commissions were appointed by the German and French
cement manufacturers for the purpose of studying the action of
concrete in sea water. The scientific and literary heirs and assigns
of the said commissions are still investigating, and sometimes
reporting.

In 1892 Michaelis, in a paper read at London to the Society
of Civil Engineers, made the following statements: "From the
chemical point of view, cements or hydraulic limes rich in silica
and as poor as possible in alumina and ferric oxide should be used,
for aluminate and ferrate of lime are not only decomposed and
softened rapidly by sea water, but they also give rise to the forma-
tion of double compounds, which in their turn destroy the co-
hesion of the mass, by producing fissures and swelling." (A. S,
C. E., Vol. 107, p. 375.)

Schuljatschenko, Eger, Rebuff at and Vicat did notable work
on the complex questions involved in their attempts to solve the
difficulty. The most convincing work was done by Le Chatelier,
who, some time previous to 1900, made several synthetic cements
and subjected these cements to the action of sea water. These
cements were manufactured in sufficient quantities to carry out
long-time tests by actual inmiersion of the test pieces. Portland
cements were manufactured, consisting of silica, alumina and
calcium, in which the alumina varied from 4.1 per cent, to 16§ per
cent. Iron oxide cements were made, in which there was no
alumina present, and consisting only of silica, ferric oxide and
calcium oxide; in these, the percentages of iron ran as high as 14
per cent. He also manufactured cements in which the oxides of
manganese, cobalt and chromium were used, as substitutes for the
ordinarily occurring alumina. In 1900, 1901 and 1902 he reported
before various scientific bodies on the action of sea water on these
special cements. As a result of his investigations, Le Chatelier took
the position that the most important, if not the only, cause of
disintegration of cement in sea water is the formation of calcium
sulpho aluminate, and that the iron oxide cements were much
more resistant to sea water than the alumina cements. (See
Reports of Paris Congress of Testing Materials, 1900; Buda
Pest Congress, 1901; and Than Industrie Zeitung, 1902, No. 11.)

Digitized by

Google

580

Brown on Marine or Iron Ore Cements.

From 1902 to the present, publications and controversies
have been frequent and confusing.

In the fall of 1909, the American Cement Engineering Com-
pany erected an experimental Portland cement plant, two views
of which are herewith presented — Fig. 1, a general view of the
entire plant, and Fig. 2, a closer view of the kiln.

The plant consists of a rotary kiln 20 ft. by 2 ft., set at an
inclination of f in. per ft., revolving at 2 turns per minute. The

PIG. 1. — INTERIOR OF LABORATORY.

kiln has a stack, 1.5 ft. in diameter by 18 ft. high; thickness of
kiln shell, J in.; lining — Harbison- Walker Special Aluminous fire
blocks, 3 in. thick. By means of counter-shafting and a variety
of different sized sprockets, several changes on the speed of the
kiln can be quickly made. The slowest speed revolves the kiln at
such a rate that the material requires somewhat over an hour to
pass from feed to discharge end. It was found that a speed which
would permit of the material passing through the kiln in one-half

Digitized by

Google

Brown on Marine or Iron Ore Cements.

581

hour was entirely satisfactory from the standpoint of quality of
clinker produced, and output. The kiln is provided with a No. 6
Rockwell fuel oil burner, complete with connections, and uses
crude oil, or distillate which is sent into the atomizing line under
a pressure from the pump of about 30 lb. per sq. in. A small
three-piston Gould power pump was first installed, but as the
valves in this pump gave considerable trouble because of residues
in the oil, this pump was replaced by a single-acting piston
power pump, of a capacity at least four times greater than was

FIG. 2. — EXPERIMENTAL KILN.

required. A steady control of feed was secured by valves in
the feed pipe, the excess oil returning to the reservoir. The
oil used was distillate.

The blower, as shown on the left of the pump, is a Rockwell
No. 1 type A positive pressure blower, provided with a relief
valve, set at 2 lb. pressure.

The reduction machinery in this plant consists of:

1 Sturtevant 8x5 laboratory roll.
1 Sturtevant 2x6 laboratory crusher.

Digitized by

Google

582 Brown on Marine or Iron Ore Cements.

1 Sturtevant sample edge grinder.

1 Abbe No. 5. 3 ft. x 3 ft. 6 in. pebble mill.

1 Abbe No. 6. 30 in. x 30 in. pebble mill.

1 Perfecticon screen.

1 Pan dryer, with sheet iron surface 5 ft. x 12 ft.

The whole plant is operated by a 12 H.P., 200 R.P.M. Backus
gasoline engine.

This plant was designed for the purpose of manufacturing
cements in sufficient quantity to make any desired practical or
laboratory test with the various classes of cement or raw ma-
terials which might be submitted for examination.

The ferruginous Miocene shell marls at Yorktown, Virginia,
furnish excellent raw material for the manufacture of a complete
series of Marine cements ranging in alumina content from 2 to
7 per cent., and in iron oxide content from 3§ to 9 per cent.
Sufficient runs were produced so that practically the entire
range of cements were secured, from a high ferric oxide cement
to a high alumina cement. Enough was made at each run so that
the characteristics of the cement in course of manufacture, the
behavior of the raw materials in process of clinkering in the kiln,
and the grindability of the raw and finished product and the
chemical and physical qualities of the cement were ascertained.

The shell marls furnished from 75 to 100 per cent, of the
raw materials entering into the various cements. When additions
were ■ necessary, clays of such analyses were selected that the
resulting cement would have the desired content of ferric oxide
and alumina.

On account of the fact that the low-alumina cements so
manufactm-ed are especially adapted for concrete miarine struc-
tures, I would recommend the name "Marine" cements, in pref-
erence to the term "Iron ore" cements, since the latter signifies
that iron ores are used in the manufacture of the cements.

The ferruginous shell marls at Yorktown, Virginia, carry
sufficient amounts of lean, ferruginous clays to make a suitable
mix for Marine cements, without the addition of tempering ma-
terials. In some of the runs made, however, in order to secure
as high ferric oxide content as desired, pyrites cinder of the fol-
lowing analysis was used :

Digitized by

Google

Brown on Marine or Iron Ore Cements.

583

Ferric oxide 79.19

Alumina 4.73

Calcium oxide 1.82

Magnesium oxide 1.35

Silica 7.72

SO3 4.95

Sulphur as sulfide 1.96

In the beginning of the ''Iron ore*' cement industry in Ger-
many, the cements as made for commercial pmposes carried less
than 1^ per cent, of alumina. These cements were so slow in
setting, and had physical characteristics so different from the
standard Portland cements, that it was difficult to convince the
buyers of their merit. It is probably on this account that the
manufacturers of ''Iron ore" cement in Germany have gradually

Table I

•

eJ'^'

n D«an.

»MO0. FluQion.

Cii^letlAiaiyni.

^ 1

i\ii

iii

1272
1813

\
im ' Mil

2iia ftsi

3IS
370

§28

193

mtil 1
tlK,l9M7JS}

SJIfll.75

0JS

1.47 so.n

1,43 97,14

! (

155

m \n.^

0,K.t2M^ZM

1 1

EJoLSajdO 64

•Two days.

increased the content of alumina, since recent importations show
the alumina content to run slightly in excess of 3 per cent.

Non-aluminous cements are practically impossible to manu-
facture, on account of the difficulties of securing raw materials
free from alumina; and even if they could be commercially manu-
factured, are less satisfactory than the cements which carry suf-
ficient amounts of alumina to render the resulting cement active
enough to meet the demands of practical constructors, who re-
quire hydraulic cements to obtain their final settmg within a
reasonable time — certainly sooner than 48 hours.

Table I shows the chemical and physical characteristics of
(first) an imported low-alumina "Iron ore** cement; (second) a
cement from the middle west; and (third) a Marine cement
made in the laboratory plant above described, the tests being

Digitized by

Google

584

Brown on Marine or Iron Ore Cements.

GO >f

-I

-a

£itS

« I

6" US

^11

si 8l i1 s| 11 11 g| g| %1 1| 11 11 1§ 11

55^5 5?'.J5 5?:r)5 5kJ ^,r| 5^.J Sr,,^ jr..^ JrJ !?.r^ ^. r^ SStg gj8 S^icg

I2;2 gg : : : : : : : : : :

•Ml^- ooeo • • ; • ; ; ; ; ; ;

S§ K^ SS : :

CM CO
o S

2?5

5»o t-'M »o» a CO — «ac »>.»H i>-r>. r-« oo» q»o obi-i ecoo e^wj £J95
•.f c« i-io —I'. OS— ^o Ot- — ^ ^?i «Bh>. ■*i^ — ■^ «p«5 «cao
z-^ Olid Qc^ o>>o 00 lo o>>75 on to cc^ eb>o oic cb>c x-^ oo^ qo<4<

52 g'o §S iS 2JS S2 gS J3f2 S« R?2 i? feS §J§ |g

«oM !»•«< ao'T a5« ooco o*"* OO'^ t--^ o6i?5 «^ oo'^ t«-* abeo *^

SS SS §g Sg 5lg SSS 2:?2 {:g ?« gi5 ffig gg gs II

•*«-i i^c^i tOM »!5?i <oesi i-w r^ec «cco t>.c«3 oecvi ir?c4 «c^ «»^ «e>»

o * o o

JrL.

M ^

M 1-1

?g s

s s

o at

*s s

oo -«

S 5 {2 8

CJ —

ao oo

^ CJ

O 00 »Q t^

i>- 00 00 t>.

!>.

«o

s ^

5 S

CSI

-^ CO

CI ^

^ r-

■^ CI

^ O -I '.t"

CO «-< OS «-«

CO O -f UJ

CI CM CO CO

Digitized by

Google

Brown on Marine or Iron Ore Cements.

585

Table III. — Effect of Accelerators on German Iron Ore Cement

No. 1272.

Accelerator Used. 'Initi